Laparoscopic or Open Inguinal Hernia Repair - Which is Best for the Patient?

Full text

(1)

(2)

(3)

(4)

(5)

(6) .

(7)

(8)

(9)

(10)

(11) . !.

(12)

(13) . "##$%$&' "#()*)$%%%)&% ++ ++,- $'.'.

(14) .

(15)

(16)

(17)

(18)

(19) !"#!$%

(20) &'

(21) %

(22)

(23)

(24) % (&

(25)

(26) &)

(27) %* +,-&

(28) .

(29) /.&, 01 , ,2

(30)

(31)

(32) 3 ' 4 5 6&&7 %

(33) & ( 8 . ,

(34)

(35) .

(36)

(37)

(38) 9:;,;", ,3/7< := !$$9:$ $, 3 ' & &

(39)

(40)

(41)

(42)

(43) ' ' ,3 & '.

(44) &

(45) . ,6 & &

(46)

(47)

(48) %& %

(49)

(50)

(51) & ,-& %

(52) &%

(53) & &% %

(54)

(55)

(56) &

(57) .

(58) & &

(59) ,3

(60)

(61)

(62) &

(63) %

(64)

(65)

(66) &

(67) & /.& *

(68) % 3 ' & 2

(69)

(70) )/*32+. ' '

(71) %

(72) !!&

(73) , 7 . <

(74) ! ; ' !$! ' ": . & ' & .

(75)

(76) &

(77)

(78) )-0(-

(79) 0 (

(80) +

(81)

(82) )2& + ,-&

(83) . % ,/

(84)

(85) . &

(86) % >

(87) %&

(88)

(89) '

(90)

(91)

(92) % % ' , 3

(93) !": ;;$ &-0( :$ &2& '

(94) .

(95)

(96) , 6 & 9?

(97) %

(98) %

(99) %

(100)

(101) . % $,! & . ",$? &-0( !,? &2& '

(102) , (

(103)

(104) .

(105) . &-0('

(106)

(107) !.1 %

(108)

(109) ' % 1 !! &

(110)

(111) %

(112) ,( &-0('

(113) & '& 1

(114) % @

(115)

(116)

(117) , A&

(118) .

(119) !!?

(120) % &-0( ! $? &2& '

(121) &%% %

(122)

(123) .

(124) , 4

(125)

(126) %

(127) -0(. &'& & %

(128) 2& .&

(129)

(130) .

(131) .

(132) &

(133) 1 , 7

(134) '

(135)

(136)

(137) > &

(138) %

(139) -0( .

(140)

(141) >

(142) . &2& , 3

(143)

(144)

(145) &-0( 2& & ' '%

(146) & , '

(147)

(148)

(149)

(150) & &

(151)

(152) %

(153) ' & ,

(154) 3 ' & -0(2&

(155) &

(156)

(157)

(158) !"# $

(159) % $

(160)

(161) &' $ . $ $"()*+,&'-$ B 01 3//<!;$!;; 3/7< := !$$9:$ $ # ### !:;")&. #CC ,1,C

(162) 8 D # ### !:;"+.

(163) We had hardly reached Öland´s shores, before we noticed that this land was completely different.. Carl von Linné, 1741.

(164)

(165) List of Papers. This thesis is based on the following papers, which will be referred to in the text by their Roman numerals. I. Eklund A, Rudberg C, Smedberg S, Enander LK, Leijonmarck CE, Österberg J, Montgomery A. Short-term result of a randomized clinical trial comparing Lichtenstein open repair with totally extraperitoneal laparoscopic inguinal hernia repair. Br J Surg.2006;93:1060-1068. II. Eklund A, Montgomery A, Rasmussen I, Sandbue R, Bergkvist L, Rudberg C. Low recurrence rate after laparoscopic (TEP) and open (Lichtenstein) inguinal hernia repair – A randomized, multicenter trial with 5 year follow-up. Ann Surg. 2009;249:33-38 III. Eklund A, Montgomery A, Bergkvist L, Rudberg C. Less chronic pain after laparoscopic than open inguinal herniorrhaphy – 5 year follow up of a RCT comparing TEP with Lichtenstein repair. Submitted. IV. Eklund A, Carlsson P, Rosenblad A, Montgomery A, Bergkvist L, Rudberg C. Cost-minimisation analysis comparing TEP and Lichtenstein for treatment of inguinal hernia. Submitted. Reprints were made with the permission of the publishers..

(166)

(167) Contents. Introduction...................................................................................................11 Historical notes.........................................................................................12 Surgical techniques ..................................................................................12 Hernia recurrence .....................................................................................14 Meshes and fixation .................................................................................17 Short-term results .....................................................................................18 Chronic pain .............................................................................................21 Economy ..................................................................................................24 Background and Aims...................................................................................27 The SMIL study group .............................................................................27 Why TEP? ................................................................................................27 Why Lichtenstein?....................................................................................27 General aim ..............................................................................................28 Specific aims ............................................................................................28 Patients and Methods ....................................................................................29 Patients .....................................................................................................29 Operations ................................................................................................29 Surgeons ...................................................................................................30 Data collection and follow-up ..................................................................30 Methods....................................................................................................30 Short-term results (Paper I) .................................................................30 Recurrence (Paper II)...........................................................................31 Chronic pain (Paper III).......................................................................31 Economic analysis (Paper IV) .............................................................31 Statistical methods ...............................................................................32 Results...........................................................................................................33 Short-term results (Paper I) ......................................................................34 Recurrence (Paper II) ...............................................................................35 Chronic pain (Paper III) ...........................................................................37 Economic analysis (Paper IV)..................................................................40 Discussion .....................................................................................................42 General remarks .......................................................................................42.

(168) Short-term results (Paper I) ......................................................................43 Recurrence (Paper II) ...............................................................................44 Chronic pain (Paper III) ...........................................................................45 Economic analysis (Paper IV)..................................................................46 Conclusions...................................................................................................48 General conclusions .................................................................................48 Specific conclusions.................................................................................48 The Future.....................................................................................................49 Sammanfattning på svenska..........................................................................51 Korttidsresultat (Delarbete I)....................................................................52 Recidiv (Delarbete II)...............................................................................52 Kronisk smärta (Delarbete III) .................................................................53 Ekonomisk analys (Delarbete IV) ............................................................53 Slutsats .....................................................................................................53 Acknowledgements.......................................................................................54 References.....................................................................................................55.

(169) Abbreviations. ASA BMI CBA CEA CI CRF CUA e-PTFE IPOM IPQ MRC na NRS ns OR QUALY RCT RR SD SHR sign. SMIL TAPP TEP VAS VRS. American Society of Anaesthesiologists Body Mass Index Cost-Benefit Analysis Cost-Effectiveness Analysis Confidence Interval Case Record Form Cost-Utility Analysis Expanded PolyTetraFluoroEthylene IntraPeritoneal Onlay Mesh Inguinal Pain Questionnaire Medical Research Council Not applicable Numerical Rating Scale Non-significant Odds Ratio Quality-Adjusted Life Years Randomised Controlled Trial Relative Risk Standard Deviation Swedish Hernia Register Significant Swedish Multicentre trial of Inguinal hernia repair by Laparoscopy TransAbdominal PrePeritoneal Totally ExtraPeritoneal Visual Analogue Scale Verbal Rating Scale.

(170)

(171) Introduction. Inguinal hernia affects both men and women but is much more common in men who comprise over 90% of operated patients1. Considering both operated and non-operated inguinal hernias, the lifetime prevalence rate is 47% for men up to and including the age of 752. Repair of an inguinal hernia is one of the most common operations among adults in the western world today. The lifetime risk of undergoing such a repair is 27% for men and 3% for women3. Not all inguinal hernias will require surgery, especially not those without any symptoms4. In spite of this, the reported annual number of operations in the Swedish hernia register (SHR)1 is approximately 17 000, which corresponds to 216/100 000 inhabitants per year. The annual number of hernia repairs in the US is approximately 800 0005. From an aetiological point of view there are two types of inguinal hernia – congenital and acquired. The term congenital hernia implies a persisting processus vaginalis with free communication between the abdominal cavity and the scrotum. In the newborn, the frequency of a patent processus vaginalis is 57-94%6. By the age of two, 37-40% still persist and half of them may develop into an inguinal hernia later in life7. An acquired inguinal hernia is generally not believed to be associated with a persistent processus vaginalis but develops in a secondary manner. Several risk factors for the development of this type of hernia have been suggested, among these various connective tissue factors. For individuals with Marfan´s and Ehlers-Danlos syndrome, cutis laxa and osteogenesis imperfecta, all of which carry an increased hernia incidence, the aetiological link to connective tissue disorders seems to be quite clear8. In order to compare the results of inguinal hernia repair, a standardised classification with respect to localisation and size is needed. Several classifications exist, of which the Nyhus classification is one of the most commonly used9 (Table 1).. 11.

(172) Table 1. Nyhus classification. Type I. Indirect hernia with normal size and structure of the internal ring. Type II. Indirect hernia with enlarged and distorted internal ring. Type IIIa. Direct hernia. Type IIIb. Indirect hernia with enlarged internal ring, encroaching on the floor of the inguinal canal. Can be accompanied by a direct hernia (pantaloon hernia). Type IIIc. Femoral hernia. Type IV. Recurrent hernia. Historical notes Some of the earliest data regarding inguinal hernia come from the Ebers papyrus (approx. 1552 BC) and the mummy of Merneptah (1224-1214 BC), which shows possible remaining signs of hernia surgery10. Over the following centuries, several documents described the anatomy and treatment of inguinal hernias with both surgical and non-surgical methods. Results were generally poor as surgical ability was fragmentary or even non-existent. Most people therefore received no treatment at all or, at best, employed the use of a truss. It was not until the second half of the 19th century, together with the introduction of anaesthesia and antiseptic techniques, that hernia surgery evolved. What can be termed modern inguinal hernia surgery started in the 1880s, with the anatomical repair introduced by Eduardo Bassini in Padua, Italy11.. Surgical techniques Hernias can only be cured by surgery. The aim is to reduce the patient’s symptoms and prevent possible negative events such as incarceration without causing new problems, e.g. chronic pain or discomfort. Many different types of repair have been described during the last century. Initially, various types of tissue repairs were used to reinforce the weak part of the abdominal wall. All these techniques implied a certain amount of tension on the suture line, an effect which may have contributed to recurrence. As a result, new “tension-free” techniques were introduced, aiming to avoid tension by the use of implant mesh. A further innovation was the development of a preperitoneal repair, first performed as an open repair and later employing various laparoscopic techniques.. 12.

(173) Presently, the repairs can be classified into three groups: - Open sutured repair: a reconstruction of the inguinal canal with tissue repair only. - Open mesh repair: a piece of mesh or mesh plug covering or obliterating the hernia defect, either on the anterior or posterior aspect of the abdominal wall. - Laparoscopic repair: laparoscopic placement of a mesh covering the myopectineal orifice on the posterior aspect of the abdominal wall. Among currently used open sutured repairs are the following: - Bassini – first described by Bassini in 188711, 12, implying an anatomical repair of the posterior wall of the inguinal canal with interrupted non-absorbable sutures. - Shouldice – based on the Bassini repair but carried out in several layers with a running suture13. This method was used as the gold standard for inguinal hernia repair in Sweden during the first part of the 1990s. - Marcy – introduced in 1871, implying a tightening of the hernia orifice with a few sutures; mainly used for small lateral hernias without weakening of the posterior wall14. Among the currently used open mesh repairs are the following: - Nyhus buttress –based on an older, open, sutured, retromuscular, preperitoneal technique, described by Nyhus in 1959. In the late 1970s this technique was modified by the addition of a preperitoneally placed mesh15. - Stoppa – another open, retromuscular, preperitoneal technique used since the 1960s. A Mersilene or Dacron mesh is placed preperitoneally through a lower midline incision covering the myopectineal orifices bilaterally16. - Lichtenstein – introduced by I. Lichtenstein in 197017, 18 and used as the gold standard for hernia repair in Sweden for the last 15 years. A polypropylene mesh is positioned on the transverse and internal oblique muscle to reinforce the posterior wall of the inguinal canal. - Plug and patch – or the Rutkow-Robbins repair described in 199319. The hernia orifice is obliterated with a cone-shaped polypropylene mesh combined with a flat polypropylene mesh. Among the laparoscopic repairs are the following: - IPOM – IntraPeritoneal Onlay Mesh, described by Filipi in 199220. The mesh is placed intraperitoneally covering the hernial orifice with a minimum of dissection.. 13.

(174) -. -. TAPP – TransAbdominal PrePeritoneal repair, first described by Arregui in 199121. A mesh is placed preperitoneally, covering the inguinal and femoral hernia orifices via a transabdominal approach. TEP – Totally ExtraPeritoneal repair, described independently in 1993 by two surgeons, McKernan and Phillips22, 23. The mesh is placed preperitoneally as in TAPP, but without entering the abdominal cavity.. Hernia recurrence Until recently, recurrence rates after inguinal hernia repair have been rather high, about 20%2, 24. When reviewing the literature, great differences are found concerning recurrence rates, most likely because of methodological disparities. From randomised controlled trials (RCT), recurrence rates of 1.6 – 10.1% after laparoscopic and 2.5 – 4.9% after Lichtenstein repair have been reported25, 26. This is in contrast with non-randomised studies where recurrence rates can be as low as 0.3% after laparoscopic and 0.1% after Lichtenstein repair17, 27. Furthermore, the follow-up time is of crucial importance. It is evident that the recurrence rate will increase with the length of follow-up and that even five years of follow-up is probably too short for a proper evaluation28, 29. Additionally, the level of attendance at the different follow-up points, i.e. the proportion of patients that are re-examined, is of importance to lower the risk of missed events30. Archibald Cochrane, the founder of the Cochrane Collaboration, aimed for a 90% attendance at all follow-up occasions31. The method of follow-up itself also affects results. Studies using a questionnaire run the risk of missing about 50% of recurrences compared with clinical examination30, 32-34. The definition of recurrence may also play a role. There are two commonly used definitions. According to Marsden, a recurrence is a weakness in the operated groin necessitating a further operation or the provision of a truss35. Shuttleworth´s definition is the presence of an expansile cough impulse in the operated groin36. With this latter definition, which includes asymptomatic patients, the recurrence rate will be higher33. Three different factors have contributed to a dramatic decrease in recurrence rates during recent years. One factor, which has revolutionised hernia surgery, is the introduction of mesh implants. Both a review and two meta-analyses37-39 have reported fewer recurrences with the use of mesh, both after open and laparoscopic repair. A second factor is the establishment of specialised hernia centres where excellent recurrence rates below 0.5% have been achieved17, 27. From non-specialised centres the reported recurrence rates are often tenfold higher1, 40. The third factor is the. 14.

(175) introduction of quality registers which have made surgeons aware of their results, thus providing an incentive to improve outcome24. There are several reasons for a recurrence. Early recurrence could be due to poor surgical technique but for later recurrences, several different factors may be of importance. A defective collagen metabolism, leading to suboptimal wound healing might play a role41, 42. In open mesh repair, most recurrences occur medially which is generally due to undersized meshes or insufficient overlap at the pubic tubercle32, 43. In laparoscopic repair, a too small mesh is the most common reason for recurrence44, 45. Scar formation around the implant leads to its substantial shrinkage46, 47, sometimes resulting in recurrence. Opinions and results concerning the need for mesh fixation are divergent48, 49. In summary, the recurrence rate in the recent largest RCTs with 400 patients are presented in Table 2. With a follow-up time between 2 – 5 years, the recurrence rates after both laparoscopic and open repair varied between 2.0 – 10.1%.. 15.

(176) Primary, recurrent, uni/bilat., men. Primary, recurrent, unilat., men, women. Primary, recurrent, uni/bilat., men, women. Primary, recurrent, uni/bilat., men, women. 538. 530. n. 200. Lichtenst.. Lichtenst.. TEP/TAPP. Open (mix). 994. 989. 507. 487. 200. TAPP. TEP. 151. 149. Open (mix). TEP. Open sutur. Open preperit. 613. TAPP. Shouldice. TAPP. Surgery. Clin.. Clin.. Clin.. Clin.. Clin.. Clin.. Evaluation. 2.0. 3.7. 5.8. 5.0. 1.0. 5.1. Follow-up (years). 86%. 72%. 65%. 48%. 96%. 86%. Attendance. sign.. 10.1% 4.9%. ns. ns. na. ns. sign. 10.0%. 4.9%. 2.5%. 1.6%. 2.0%. 2.0%. 3.0%. 6.7%. 6.6%. Recurrence. The SCUR trial has three arms but the actual number of patients in each arm is not reported. Differences in recurrence rates between groups could therefore not be evaluated. The MRC trial which initially had 486patients in the laparoscopic and 460 in the open arm is similarly difficult to evaluate as only a subgroup analysis of parts of the material has been published.. Jan 1999 – Nov 2001. Veterans26. Feb 1994 – Jun 1995. Coala53. May 1995 – Dec 1996. Whipps Cross25. Jan 1994 – March 1997. MRC52. Nov 1993 – Jun 1996. SCUR51. Primary, recurrent, unilat., men. Primary, unilat., men. SMIL-150. Feb1993 – March 1996. Type of hernia. Trial. Table 2. Recent RCTs comparing recurrence rates after laparoscopic and open repair for inguinal hernia.

(177) Meshes and fixation For more than a century, different materials have been tested in the search for an ideal prosthetic substance to reinforce the groin and facilitate hernia repair. Theodor Billroth stated: “if we could artificially produce tissues of the density and toughness of fascia and tendon, the secret for the radical cure of hernia would be discovered”54. Among those materials tested have been silver filigree in the late 19th century55, homologic materials such as tendon from kangaroo, whale, ox and deer, and autologic materials such as fascia lata and the aponeurosis of the rectus muscle. Biological materials have the disadvantage of undergoing complete phagocytotic degeneration over time10. Metal meshes made of tantalum and stainless steel have been abandoned due to fragmentation, fracturing, sinus formations and recurrences10. One early synthetic material used for hernia repair was nylon. This was first introduced by French surgeons in Marseille in 1944 with a method similar to the Lichtenstein technique56. Other synthetic materials such as polyvinyl alcohol, Silastic, Teflon and carbon fibre have been tested but are no longer in use10. During the last two decades of the 20th century, three synthetic meshes have mainly been used: polypropylene, polyester and expanded polytetrafluoroethylene (e-PTFE). Polypropylene (Marlex®, Prolene®) was synthesised in 1954 and has been used for hernia repair since 195857. Polyester (Mersilene®) was synthesised in 1939 and introduced into hernia repair in 195654. Patented in 1976, e-PTFE (Gore-Tex®) has mainly been used for vascular prosthesis but to some extent also for hernia repair. The most commonly used material today is polypropylene, followed by polyester. The mesh technique has progressed from small to large implants. For laparoscopic repair, mesh size has increased in order to avoid recurrence44, 45, 58 and to compensate for possible shrinkage. Presently mesh size is approximately 13 by 15 cm to overlap the inguinal and femoral regions in all directions. For the Lichtenstein repair, mesh size has increased accordingly. Early meshes were 5 by 10 cm59 or even smaller32. The recommended size now is 7 by 15 cm. The mesh should overlap the pubic tubercle by approximately 2 cm medially, 3-4 cm above the Hesselbach´s triangle superiorly and by 5-6 cm laterally to the internal ring60. The most commonly used and studied mesh, polypropylene, has been developed further during the last decade. Meshes are now categorised as heavyweight or lightweight. There is no clear definition for these two, but a weight below 50g/m2 is generally referred to as lightweight. The concept of lightweight meshes is that less foreign material is deposited in the wound, thus reducing any chronic inflammatory response. One effect of this is a reduction of the scar plate formed, possibly leading to a reduced risk of 17.

(178) chronic pain. This was indicated by some studies after Lichtenstein repair61-63, but data are somewhat inconsistent64. To lower the weight of the meshes, pore size and fibre thickness have been modified. A pore size of at least 75 microns is an important property of the mesh, facilitating the ingrowth of vessels and collagen fibres as well as the admission of macrophages46. The pore size also influences the character and amount of adhesions between the mesh and intestines65. Some lightweight meshes are manufactured with an absorbable component in addition to polypropylene to make them easier to handle during the operation66. There are two commonly used absorbable components, polyglactin and polyglecaprone. The impact of the mesh coating or whether it is multi- or monofilamentous presently remains unclear67. Whether a mesh should be fixed or not is debatable. In the original Lichtenstein repair the mesh was fixed with both interrupted stitches and a running suture of non-absorbable material. The need for this has been questioned due to the risk of chronic pain. The use of an absorbable suture has not resulted in less chronic pain68. Human fibrin glue for fixation seems effective69, but the effect on chronic pain is still unclear. An international RCT on this issue is currently ongoing70 and results are awaited. In accordance with the initial description of the TEP repair, fixation of the mesh with staples or titanium screws was commonly used. It has been discussed whether these staples could cause chronic pain due to nerve injury, injuries to the deep muscle layers or to the pubic area71. To overcome this, the mesh can be left in place without fixation or it can be fixed with human fibrin glue. Both methods appear effective but their impact on chronic pain is unclear72-75. The most novel method for fixation is a “self-gripping” mesh with micro hooks made of polylactic acid. This mesh is also semi-absorbable (polylactic acid) and is therefore also considered to be a lightweight mesh. This material has so far only been used in one published prospective trial including 52 patients with 70 hernias. The mesh was placed with an open technique in an anterior position. After 24 months, no recurrences were found and only one patient reported discomfort76. There are two RCTs currently underway comparing this new implant with a sutured lightweight mesh77.. Short-term results With the introduction of the laparoscopic technique it was of interest to study the early postoperative period to see if there were any patient benefits in comparison with conventional open techniques. One factor of major importance in this respect is postoperative pain which is considered to reflect the amount of surgical trauma caused by the repair. 18.

(179) It is also the main determinant for postoperative recovery, including length of sick leave. When interpreting postoperative pain, one must consider that several variables correlate with its degree, such as patient age, the amount of preoperative pain, gender, length of incision and type of operation78, 79. In order to evaluate short-term results and time to full recovery, several methods can be used. One objective way to evaluate impairment due to surgical trauma is to use functional tests resembling ordinary daily activities. Among these are: the treadmill test, raising straight legs, curled sit-ups, squatting, raising from a bed and climbing stairs. These tests are generally all in favour of laparoscopically operated patients one week after surgery80-82. It is, however, not common to use functional tests in hernia trials. Instead, a proxy such as “time to recovery”, “performs usual activities” or “takes part in social activities” are used. The amount of time to achieve this is stated by the patient. This makes it difficult to compare the results from different studies, even if the proxy is well-defined. Several reports, however, indicate that the postoperative course for patients operated with the laparoscopic technique is more favourable37, 39. Length of sick leave is another way to measure physical impairment after surgery. Over the last few decades, there have been tremendous changes in recommendations for the length of sick leave. A duration of seven to eight weeks was normal 30 years ago, four to six weeks 15 years ago and one week five years ago83, 84. The interpretation of the length of sick leave has its pitfalls, however, since it is influenced by several factors such as patient expectations, personality, information from friends and colleagues, the type of compensation system, if there is pain or other complications from the wound, and recommendations from the medical profession itself83-89. Metaanalyses may reduce doubts arising from single studies, and the most recent one reports on an earlier return to work after laparoscopic compared with open repair39. To summarise, short-term results from the largest recent RCTs with 400 patients are presented in Table 3. Convalescence is more favourable after laparoscopic repair regarding postoperative pain, sick leave and time to full recovery.. 19.

(180) Primary, recurrent, uni/bilat., men. Primary, recurrent, unilat., men, women. Primary, recurrent, uni/bilat., men, women. Primary, recurrent, uni/bilat., men, women. Lichtenst.. Open (mix) TEP/TAPP 994. 507 989. 487. 200. Lichtenst. TEP. 453 200. 462. 613. 524. 518. n. Open (mix) TAPP. Shouldice TAPP Open, prep Open sutur TEP/TAPP. TAPP. Surgery. Clin.. Clin.. Clin. 1 w. and Enquiry 12 w. Clin.. Clin.. Clin.. Evaluation. --. --. 90%. 89 and 74%. 96%. --. Attendance. Less pain sign.. Less pain sign.. Less pain sign.. Less pain sign.. Less pain sign.. Less pain sign.. Postop pain. 15. 21 13. 14. 18. 42 11. 14 15 18 18 28. 10. sign.. sign.. sign.. sign.. sign.. sign.. Sick leave (days). 36 Shorter. 24. 21. 18 24 26 14. --. --. sign.. sign.. sign.. sign.. Time to recover (days). The SCUR trial reported less postoperative pain for the TAPP group compared with the open suture group. The same was true for sick leave. Concerning time to recovery, the TAPP group was compared with both the open preperitoneal and the open suture group. In the MRC trial, postoperative pain is from Scott et al91. Time to recovery was expressed as “able to enjoy usual interests and hobbies”. In the Whipps Cross trial, sick leave concerns sedentary work and was expressed as a hazard ratio (95% CI): 0.41 (0.23 to 0.75). In the Coala trial, time to recovery was expressed as “time to resumption of athletic activities”. In the Veterans trial, sick leave is from Hynes et al94.. Jan 1999 – Nov 2001. Veterans26, 94. Feb 1994 – Jun 1995. Coala93. May 1995 – Dec 1996. Whipps Cross92. Jan 1994 – March 1997. MRC90, 91. Nov 1993 – Jun 1996. SCUR51. Primary, recurrent, unilat., men. Primary, unilat., men. SMIL-180. Feb1993 – March 1996. Type of hernia. Trial. Table 3. Recent RCTs comparing short-term results after laparoscopic and open repair for inguinal hernia.

(181) Chronic pain Chronic pain is considered to be the most important outcome measure of inguinal hernia repair since it is a common and long-lasting problem for the patient, while recurrence rates are low70. The definition of chronic pain is pain or discomfort which lasts for more than three months postoperatively95. The incidence of chronic pain after herniorrhaphy varies widely. It lies between 0 and 75% after open mesh and 0 and 29% after laparoscopic repair71, 96, 97. The frequency of pain that affects daily activities is reported to be in the range of 5-6%98, 99. Several authors report mesh repair to result in less chronic pain than non-mesh, and laparoscopic less than open mesh repair37, 39, 100. Many tools are available to investigate the frequency and severity of chronic pain. One-dimensional methods used are: the Visual Analogue Scale (VAS), the Verbal Rating Scale (VRS) and the Numerical Rating Scale (NRS). With questionnaires such as the McGill Pain Questionnaire and Inguinal Pain Questionnaire (IPQ), a multidimensional assessment of the sensory, affective and functional effects of pain can be investigated101, 102. A multidimensional questionnaire that includes functional aspects is of greater value than one-dimensional tests, due to its clinical implications. The recently introduced IPQ is specifically designed for the investigation of chronic pain after inguinal hernia repair101. Pain has been thought to be either neuropathic or nociceptive in nature, or a mixture of both. In one study with an extensive neurophysiological investigation of 46 patients with chronic pain, all patients were considered to have pain due to nerve injury. This pain primarily originated from deeper rather than cutaneous structures103. Another author reported a neuropathic origin of pain in 68% of patients104. These findings are of interest for treatment strategies and prophylaxis. There are several options for the treatment of chronic pain which can be divided into surgical and non-surgical. Expectancy is the primary non-surgical treatment. There is some evidence of a “burn-out” effect and of spontaneous improvement over time100, 104. Pharmacological treatment, physiotherapy or referral to a pain clinic are other options that may be considered after the exclusion of recurrence or other causes of chronic pain. For those patients with substantial pain that do not respond adequately to non-surgical treatment, surgery might be an option. The commonly used surgical method is neurectomy of one to all three nerves in the inguinal region. In some cases this is combined with mesh removal105, 106. Results after surgical treatment for chronic pain are, however, questionable107. Preventive measures are therefore of interest. Methods are either careful surgical technique with dissection and preservation of the nerves in the inguinal region or the prophylactic division of the same nerves. It would appear to be of advantage, however, to preserve the nerves108-111. 21.

(182) Furthermore, the use of mesh has been suggested to result in less chronic pain compared with no mesh100. In this respect, laparoscopic repair might be slightly better than an open approach39, 96, 100. Mesh weight and type of fixation are two other factors that are currently under investigation. This has already been discussed under the Meshes and fixation section of this thesis. Patients themselves can also carry risk factors for chronic pain. Young age, female sex, preoperative pain and operation for a recurrent hernia are known risk factors96, 98, 100. Pain-related sexual dysfunction is a more novel and less investigated type of complication of hernia repair. One author reported this type of pain in 20% of patients, resulting in clinically significant problems in 3% of men aged from 18 to 40 years112. To summarise, the results on chronic pain from the largest recent RCTs with 400 patients are presented in Table 4. The number of patients suffering from chronic pain is reported to range from 0 to 20%. Laparoscopic repair seems to render fewer problems.. 22.

(183) Primary, recurrent, uni/bilat., men. Primary, recurrent, unilat., men, women. Primary, recurrent, bi/unilat., men, women. Primary, recurrent, bi/unilat., men, women. 200. Lichten.. Lichten.. TEP/TAPP. Open (mix). 993. 989. 507. 487. 200. TAPP. TEP. 453. 462. 613. 538. 530. n. Open (mix). TEP/TAPP. Open sutur. Preperit.. TAPP. Should.. TAPP. Technique. Clin.. Clin.. Clin.. Enquiry. Clin.. Clin.. Evaluation. 2.0. 3.7. 5.8. 5.0. 1.0. 5.2. Follow-up (years). 86%. 72%. 65%. 61%. 96%. 81%. Attendance. 9%. 6%. --. 15%. 5%. 20.1%. 18.1%. --. 11.4%. 8.5%. sign.. sign.. ns. ns. Chronic pain. In the Whipps Cross trial, patients with groin and testicular pain are found in the same group. The Coala and SCUR trials gave no conclusive information concerning the number of patients with chronic pain.. Jan 1999 – Nov 2001. Veterans115. Feb 1994 – Jun 1995. Coala53. May 1995 – Dec 1996. Whipps Cross25. Jan 1994 – March 1997. MRC114. Nov 1993 – Jun 1996. SCUR51. Primary, recurrent, unilat., men. Primary, unilat., men. SMIL-1113. Feb1993 – March 1996. Type of hernia. Trial. Table 4. Recent RCTs comparing chronic pain after laparoscopic and open repair for inguinal hernia.

(184) Economy Politicians and other individuals with decisive positions in the health care system need data from economic analyses when making decisions about new medical methods. Several types of analyses for economic evaluation are available. The four most commonly used are116: Cost-minimisation analysis, which is the simplest form of economic evaluation. It is criticised for lacking completeness117 as it solely compares costs for the methods in question. This method is recommended when the primary endpoint is considered to be equal between the groups. Cost-effectiveness analysis (CEA) determines the trade-off between cost and health outcome. It is used when the endpoints are considered to differ. The consequence on health outcome is expressed in terms of non-monetary units that describe the desired objective such as lives saved (i.e. life years gained), number of hospitalisations avoided or cases (recurrences) prevented. Cost-utility analysis (CUA) is used when quality of life is an important factor to consider. CUA is similar to CEA regarding cost but different in respect to outcome, which is often expressed in terms of quality-adjusted life years (QUALY). There is an increasing use of QUALYs. To use this analysis, the investigator often relies on existing utility questionnaires. Cost-benefit analysis (CBA) is used when the benefit can be transformed into a monetary value which is therefore easily comparable with the cost. It is possible to determine whether a new intervention gives a total benefit to society - with benefits exceeding costs - which is a fuller analysis compared with a CEA. The management of inguinal hernia has undergone tremendous changes over the last few decades. Specifically, three things have changed: an increasing number of operations are performed as day surgery, recommendations for sick leave and convalescence times are shorter, and there has been a substantial decrease in recurrence rates. All of these factors have contributed to decreasing costs for inguinal hernia repair. When laparoscopic methods were introduced for inguinal hernia repair, costs were expected to be higher for the hospital when compared with the open technique. This situation has been confirmed92, 118. Over a more extended time perspective, including the costs for early complications, recurrences and sick leave, laparoscopy tends to be equally or even less expensive than open repair94, 119-122. There is, however, a shortage of longterm data concerning total societal and hospital costs. In summary, the results from an economic point of view in the largest RCTs with 400 patients are presented in Table 5. Hospital costs are higher for laparoscopic repair. When costs for sick leave, recurrences and complications are included, the outcome is similar to open repair.. 24.

(185) Primary, recurrent, uni/bilat., men. Primary, recurrent, unilat., men, women. Primary, recurrent, uni/bilat., men, women. Primary, recurrent, uni/bilat., men, women. Lichten.. Open (mix) TEP/TAPP 788. 139 763. 134. 174. Lichten. TEP. 460 176. 486. 613. 538. Open (mix) TAPP. TEP/TAPP. Open sutur. Preperit.. Should. TAPP. 530. n. 2.0. 1.0. 0.25. 1.0. 1.0. 5.1. Follow-up (years). 706. 556 1452. 970. 271. 1111 734. 0 1422. 51. 875. --. --. --. --. sign.. --. Hospital. 1835. 1968. 735. 778. 1467. --. --. --. --. Hospital and community. --. --. 8159. 1872 8742. 1973. 469. 1271 1047. 1792. --. --. ns. --. --. --. Total cost including recurrences and complications. Cross trial, 50 patients (24 laparoscopic and 26 open) were omitted from the analysis due to incomplete data and, furthermore, there was no test of significance between groups.. In the SCUR trial, sutured repair was set to nil and compared with two other methods and their additional costs. In the Whipps. Jan 1999 – Nov 2001. Veterans94. Feb 1994 – Jun 1995. Coala123. May 1995 – Dec 1996. Whipps Cross92. Jan 1994 – March 1997. MRC118. Nov 1993 – Jun 1996. SCUR51. Primary, recurrent, unilat., men. TAPP. Primary, unilat., men. SMIL-1. Feb 1993 – March 1996. Technique. Type of hernia. Trial. Table 5. Recent RCTs comparing costs in € (2007) for laparoscopic and open repair for inguinal hernia..

(186)

(187) Background and Aims. The SMIL study group In 1993, a study group was formed to evaluate the possible benefits of laparoscopic hernia repair compared with open techniques. The aim was to conduct well-designed multicentric, large RCTs with general surgeons in their routine clinical setting. The first trial, the Swedish Multicentre trial of Inguinal hernia repair by Laparoscopy (SMIL-I) was started in February 1993, and compared TAPP with Shouldice for primary inguinal hernia, and TAPP with Lichtenstein for recurrent inguinal hernia. Since the inclusion of patients into this trial was swift and more centres were added to the group, a second RCT was launched in November 1996 - the SMIL-II trial. In this trial, TEP was compared with Lichtenstein for primary inguinal hernias. The primary endpoint for both studies was recurrence after five years.. Why TEP? Several laparoscopic techniques have been introduced for hernia repair. In the earliest report by Ger124, stainless steel clips were used to close the hernial orifice. Later on, mesh was introduced to cover the defect in the abdominal wall, either in an intra- or extraperitoneal fashion. Intraperitoneal onlay mesh (IPOM) was soon abandoned due to an unacceptably high recurrence rate and adhesion formation sometimes leading to mechanical bowel obstruction21, 125. To prevent this complication, the technique was further developed with placement of the mesh extraperitoneally which could be made with either the TAPP or TEP technique. The TEP technique is somewhat more difficult to learn, but is considered to have less complications126, 127. The latter was an important factor when choosing this method for the present trial.. Why Lichtenstein? The laparoscopic TEP repair is a tension-free method, and therefore it was logical to compare it with an open tension-free method. The Lichtenstein technique was popularised in Sweden during the mid 1990s and became the 27.

(188) “gold standard” for open hernia repair. The reasons for this were the low recurrence rate and relative ease with which it was mastered by the average surgeon17, 128, 129.. General aim The general aim was to investigate if general surgeons in routine surgical practice could reproduce the excellent results reported by hernia experts and specialised hernia centres.. Specific aims -. -. 28. To compare TEP with Lichtenstein repair concerning short-term results; namely postoperative pain, sick leave and time to resumption of normal physical activities. To compare TEP with Lichtenstein repair concerning recurrence rates at five years. To evaluate risk factors for recurrence after TEP and Lichtenstein repair. To compare TEP with Lichtenstein repair concerning chronic pain at one, two, three and five years. To evaluate risk factors for chronic pain after TEP and Lichtenstein repair. To compare TEP with Lichtenstein repair concerning hospital and community costs, including recurrence and complications, after five years of follow-up..

(189) Patients and Methods. This thesis is based on one large multicentre RCT. Four different aspects of this trial have been investigated in four individual papers (Papers I to IV). There were 11 participating centres: two university, six regional and three county hospitals. The trial was approved by the local ethics committees and was planned according to the intention to treat principle. Randomisation of patients took place between November 1996 and August 2000.. Patients Men aged 30 to 70 years with a primary unilateral inguinal hernia were eligible for inclusion. Exclusion criteria were recurrent, scrotal or bilateral hernia, mental disorders, language difficulties and an ASA (American Society of Anaesthesiologists) grade higher than III, or who were otherwise unfit for general anaesthesia.. Operations Laparoscopic repair was performed as TEP130. A space was created preperitoneally in the inguinal region with a balloon dissector (PBD® 100; Origin Medsystem, Menlo Park, California, USA). After inversion or division of the hernial sac, a 12 by 15 cm polypropylene mesh (Atrium Medical, Hudson, New Hampshire, USA) was introduced and positioned to cover all three possible hernial orifices (lateral, medial and femoral). The mesh was fixed with titanium screws (Origin-Tacker® 5mm; Origin Medsystem). The open repair was performed according to Lichtenstein17. The hernial sac was invaginated or resected. The posterior wall was reinforced by a 7.5 by 12-15 cm polypropylene mesh (Atrium Medical, Hudson, New Hampshire, USA) which was anchored with a 2/0 polypropylene running suture to the inguinal ligament and interrupted stitches in the medial and superior part of the mesh. Irrespective of the operative method, patients were informed to resume normal daily activities and return to work as soon as possible.. 29.

(190) Surgeons All surgeons were board certified, 22 in the TEP and 26 in the Lichtenstein group. Both groups were experienced with their respective technique and each surgeon was expected to contribute a minimum of 25 patients to the trial. No surgeon was to perform both methods.. Data collection and follow-up Demographics and a functional test were recorded before surgery. Operative and postoperative variables were registered, including complications, time in hospital, functional test results one week after surgery, sick leave and time to complete recovery. A visual analogue scale (VAS; 0-100 mm) was used by the patients to measure postoperative pain. Clinical controls were performed by an independent observer one and 12 weeks, and one and five years after surgery. At two and three years a questionnaire was used. This consisted of two questions: 1. possible hernia recurrence? (yes/no) and 2. complaints after surgery? (yes/no). If complaints after surgery were reported, patients were asked to elaborate further.. Methods Short-term results (Paper I) Close postoperative follow-up was carried out for 12 weeks in order to detect all possible complications and monitor postoperative pain and recovery. Complications occurring intraoperatively and up to the time of discharge were registered in the patient’s case record form (CRF). Postoperative pain was assessed by the patient using a VAS once daily during the first week and once a week during the following 11 weeks. The patients also recorded the daily amount of oral analgesics during the first week after surgery. At the first check-up one week after surgery, the independent observer assessed the patient’s functional status, consisting of rising from a bed, climbing stairs and squatting. Unrestricted activity was graded as 1, some restriction as 2 and great difficulty as 3 points. The test was combined into a functional index score (range 3-9 points). This functional test was an objective test of the patient’s capacity for ordinary daily activities.. 30.

(191) Recurrence (Paper II) Clinical examination was performed by an independent observer at one and five years after operation to reveal possible recurrence and other complaints. After two and three years, a written questionnaire was sent to the patients. A recurrence was defined as the presence of an expansile cough impulse36. If the clinical examination was inconclusive, a herniography was performed. Possible risk factors for recurrence were also recorded.. Chronic pain (Paper III) All complaints recorded in the questionnaire from the different follow-up occasions, together with those complaints recorded by the independent observer at one and five years, were judged and categorised by two independent and blinded observers. The resulting categories were: no complaints, mild pain (occasional discomfort or pain not interfering with daily activities), moderate pain (discomfort or pain occasionally interfering with daily activities) and severe pain (discomfort or pain interfering with daily activities). A consensus discussion was held in the case of differences of opinion. A further analysis was performed on the group of patients who were categorised as having moderate or severe pain at any of the check-up points from one year onwards. This group of patients received the validated IPQ98, 101, consisting of 18 questions. This questionnaire was designed to ascertain whether the complaints changed over time and whether the discomfort was disabling for the patient. A further five questions concerning medical care and sick leave due to the operation were added to the questionnaire. Possible risk factors for chronic pain were extracted from the CRF and were studied for the whole group, as well as for the TEP and Lichtenstein groups, respectively.. Economic analysis (Paper IV) Different aspects of costs in connection with hernia repair were included in the analysis. Operative time, sick leave and time to full recovery were derived from the CRF. Costs for anaesthesia, recovery room, laparoscopic equipment and participating surgeons were calculated from a subgroup of patients. These were selected in a random manner from one medium-sized hospital which was judged to represent an average institution between the participating small county hospitals and large university clinics. Figures for anaesthesia and recovery room included overhead costs. All costs were categorised as hospital and community costs and expressed in the average value of the Euro in 2007. 31.

(192) Statistical methods The sample size was calculated such that a 5% difference ( = 0.05) in recurrence rate at five years between the groups could be detected with a power of 0.9. This required a sample size of 1200 patients. Considering an estimated drop-out rate of 10%, we aimed at including 1320 patients. Continuous data were compared using Student’s t test. Non-metric data were compared using the Mann-Whitney U test, and frequencies with Pearson’s 2 and Fisher’s exact test. Risk calculations were performed with Cox regression for relative risk and conditional logistic regression for odds ratio. Specific statistical analysis in Paper II: A Z-test for heterogeneity was used for an individual surgeon. Specific statistical analysis in Paper III: Kappa statistics tested for concordance between the observers. Bootstrap testing was used for multiple dependent pairs. Specific statistical analysis in Paper IV: A Z-test was used for determining the mean values for two independent samples.. 32.

(193) Results. A total of 1512 patients were randomised. For various reasons, 142 patients (9.4%) were excluded before surgery (Fig 1). Overall, 1370 patients were operated, 665 in the TEP and 705 in the Lichtenstein group, between December 1996 and May 2001.. Randomised Randomised patients patients 1512 1512. Excluded (n=93) Refused participation paricipation (n=16) Randomisation error (n=38) Surgeon not available (n=19) Medical reasons (n=14) Other reasons (n=6). TEP. Lichtenstein. 758. 754. Operated. Operated. 665. 705. 1 week 86.2% (573/665). 3 months 86.9% (578/665). Excluded (n=49) Refused participation paricipation (n=11) Randomisation error (n=14) Surgeon not available (n=14) Medical reasons (n=4) Other reasons (n=6). 1 week 87.0% (613/705). 3 months 87.4% (616/705). 5 recurrences. 1 year 82.7% (546/660). 1 year 81.8% (577/705). 7 recurrences. 3 recurrences. 2 years Year 83.5% (545/653). 3 years Year 84.8% (554/653). 2 years Year 82.8% (581/702). 3 years Year 83.9% (589/702) 2 recurrences. 5 years Year 94.3% (616/653). Not followed -up up Recurrences (n=12) Deaths (n=16) Moved abroad (n=4) Refuses follow -up (n=16) Refused Lost to follow -up (n=1). 5 years Year 94.1% (659/700). Not followed -up up Recurrences (n=5) Deaths (n=9) Moved abroad (n=2) Refuses follow -up (n=27) Refused Lost to follow -up (n=3). Figure 1. Flow chart.. 33.

(194) Short-term results (Paper I) The median operating time was 55 minutes in both groups, with a range of 12-180 min in the TEP and 20-145 min in the Lichtenstein group. Operation time in the TEP group decreased continuously with an increasing number of operations (r2=0.82). There was also a trend of less gas leakage from the extraperitoneal space into the intraperitoneal cavity during the trial. There were more intraoperative complications in the TEP group. Out of a total of 32 patients, 12 experienced complications due to technical problems with the laparoscopic equipment and another 12 operations had to be converted - eight to open repair and four to TAPP (Table 6). The total number of complications at discharge, one week and three months after surgery was equal between the groups. The postoperative course was easier and swifter for the TEP group, with less postoperative pain, five days less sick leave (P<0.001) and 11 days shorter time (P<0.001) to full recovery. During short-term follow-up, eight (1.2%) major complications occurred in the TEP group (operative complications, reoperations and readmissions) and three (0.4%) in the Lichtenstein group (P=0.030). Table 6. Short-term results. TEP n=665. %. LICHTENSTEIN n=705. %. P. Intraoperative. 32. 4.8. 15. 2.1. 0.006. At discharge. 49. 7.4. 72. 10.2. 0.065. At one week. 99. 17.3. 107. 17.5. 0.889. At three months. 83. 14.1. 101. 16.3. 0.277. Postoperative pain. Median. Range. Median. Range. VAS (mm), combined days 1,2,3,5 and 7. 105. 0-440. 175. 0-440. <0.001. Analgesics, combined days 1,2,3,5 and 7. 4. 0-40. 14. 0-48. <0.001. Sick leave, days. 7. 0-77. 12. 0-55. <0.001. Full recovery, days. 20. 0-179. 31. 0-163. <0.001. Complications. 34.

(195) Recurrence (Paper II) After a median follow-up of 5.1 (4.4-9.1) years, the cumulative recurrence rate was 3.5% (n=21) in the TEP and 1.2% (n=7) in the Lichtenstein group (P=0.008) (Fig 2). In the TEP group, one surgeon had a disproportionately high number of recurrences (7/22) and was considered to be an outlier when tested for heterogeneity (P<0.001). A subgroup analysis of the TEP group excluding this surgeon lowered the cumulative recurrence rate to 2.4%, which did not differ statistically from the Lichtenstein group (P=0.109). A surgeon age of 45 years or more and an operation time over 40 minutes decreased the risk of recurrence in the TEP group, (RR 0.24, CI 0.08-0.67 and RR 0.24, CI 0.08-0.71) (Table 7). In the Lichtenstein group, only smoking was found to be a risk factor for recurrence, (RR 12.52, CI 2.39-65.55).. 4. %. 3,5. 3 2,5 TEP Lichtenstein. 2 1,5 1 0,5. 0. 1. 2. 3. 5. Years. Figure 2. Cumulative recurrence rates for the TEP and Lichtenstein repair, respectively.. 35.

(196) Table 7. RR and 95% CI from a multivariable analysis of risk factors for recurrence. TEP RR Bruises at one week. No. Reference. Yes. 2.05. < 40 min. Reference. > 40 min. 0.24. < 45 years. Reference. > 45 years. 0.24. < 180 mm. Reference. > 180 mm. 3.80. < 25. Reference. > 25. 2.47. No. Reference. Yes. 2.99. No. Reference. Yes. 0.27. Lichtenstein 95% CI. RR. 95% CI. Reference 0.77 - 5.39. 0.97. 0.18 - 5.32. Reference. Operating time 0.08 - 0.71. 0.52. 0.89 - 2.99. Reference. Surgeon age Sum of VAS week 1-12. 0.08 - 0.67. 1.28. 0.28 - 5.97. Reference 1.44 - 9.99. 1.55. 0.33 - 7.15. Reference. BMI Intraoperative complication. 0.86 - 7.14. 1.37. 0.28 - 6.67. Reference 0.86 - 10.40. 5.87. 0.58 - 59.31. Reference. Smoking habit. 36. 0.03 – 2.11. 12.52. 2.39 – 65.55.

(197) Chronic pain (Paper III) The risk of being classified as having chronic pain was approximately twice as high in the Lichtenstein as in the TEP group at all follow-up points (Fig 3). For mild chronic pain, this risk was more than twice as high for the Lichtenstein group (Fig 4). The number of patients classified as having severe pain was low in both groups. Patients with severe and moderate pain were therefore merged together, creating the moderate/severe pain group. In this combined group, fewer patients in the TEP than in the Lichtenstein group had chronic pain at all follow-up points (Fig 5). The number of patients with moderate/severe pain after the Lichtenstein repair decreased over time (P=0.001). In the subgroup of patients who received the IPQ, 90% responded to the questionnaire. After a median follow-up time of 9.4 (6.7-10.8) years, 60% of patients (72/121) no longer reported any pain. Univariable analyses on risk factors for TEP and Lichtenstein are presented in Table 8. Multivariable analyses for risk factors in the whole group (TEP and Lichtenstein combined) are presented in Table 9. By multivariable analysis, risk factors for chronic pain in the TEP group were impairment in the physical test after one week, a recovery time longer than the median and body weight below the 3rd quartile. For the Lichtenstein group, it was pronounced postoperative pain. These multivariable analyses for each group are not shown in the tables. %. 30 P<0.001. 25 Proportion of patients (%). P<0.001 P<0.001. 20. P<0.001. TEP 15. Lichtenstein. 10 5 0. 1. 2. 3. 5. Years. Figure 3. Proportion of patients reporting chronic pain (mild, moderate and severe) 1-5 years after operation.. 37.

(198) 20. % P<0.001. 18 P<0.001 Proportion of patients (%). 16. P<0.001. P =0.001. 14 12 TEP. 10. Lichtenstein. 8 6 4 2 0. 1. 2. 3. 5. Years. Figure 4. Proportion of patients reporting mild chronic pain 1-5 years after operation.. 10. % P <0.001. 9. Proportion of patients (%). 8. P =0.001. 7. P =0.079. 6 TEP. 5 P =0.034. 4. Lichtenstein. P =0.092. 3 2 1 0. 1. 2. 3. 5. Years. Figure 5. Proportion of patients reporting moderate/severe chronic pain 1-5 years after operation.. 38.

(199) 27/276(9.8). 14/286(4.9). < Median. > Median. 15/131(11.5). 36/415(8.7). > 45. > Quartile. 18/149(12.1). Impaired. 33/432(7.6). 34/436(7.8). Not impaired. < 3 Quartile. 16/116(13.8). Restricted. 20/200(10.0). 41/492(8.3). Unrestricted. < 45. 50/454(11.0). 7/157(4.5). > 3 Quartile. < 3 Quartile. 37/307(12.1). 20/308(6.5). < Median. > Median. 2.11. 1.0. 1.56. 1.0. 1.30. 1.0. 1.62. 1.0. 1.76. 1.0. 2.65. 1.0. 1.97. 1.0. OR. TEP. 1.08 – 4.11. Reference. 0.82 – 2.98. Reference. 0.74 – 2.28. Reference. 0.89 – 2.97. Reference. 0.95 – 3.26. Reference. 1.18 – 5.98. Reference. 1.12 – 3.48. Reference. 95% CI. *Sum of all registered VAS values during the first 12 weeks of postoperative follow-up.. Time to recovery. Sum of VAS week 1-12*. Surgeon age. Physical test Pre minus Post. Physical test preop. BMI. Patient age. Pain at 5 years cases/total (%). Table 8. Univariable analysis of risk factors for chronic pain.. 53/292(18.2). 47/292(16.1). 37/144(25.7). 71/444(16.0). 70/311(22.5). 54/348(15.5). 59/253(23.3). 59/383(15.4). 30/106(28.3). 93/544(17.1). 92/490(18.8). 32/163(19.6). 74/329(22.5). 50/330(15.2). Pain at 5 years cases/total (%). 1.16. 1.0. 1.82. 1.0. 1.58. 1.0. 1.67. 1.0. 1.91. 1.0. 0.95. 1.0. 1.62. 1.0. OR. Lichtenstein. 0.75 – 1.77. Reference. 1.16 – 2.85. Reference. 1.07 – 2.34. Reference. 1.12 – 2.50. Reference. 1.19 – 3.09. Reference. 0.60 – 1.48. Reference. 1.09 – 2.42. Reference. 95% CI. 39.

(200) Table 9. Multivariable analysis of risk factors for chronic pain, both groups combined. Multivariable Pain at 5 years cases/total (%). OR. 95% CI. > Median. 60/640. (9.4 ). 1.0. Reference. < Median. 112/635. (17.6 ). 1.57. 1.07 – 2.30. Physical test preop. Unrestricted. 134/1036. (12.9 ). 1.0. Reference. Restricted. 46/222. (20.1 ). 1.86. 1.18 – 2.93. Physical test pre minus post. Not impaired. 93/819. (11.4 ). 1.0. Reference. Impaired. 77/402. (19.2 ). 1.59. 1.07 – 2.37. > 45. 90/763. (11.8 ). 1.0. Reference. < 45. 90/511. (17.6 ). 1.55. 1.06 – 2.27. < Median. 50/570. (8.8 ). 1.0. Reference. > Median. 91/576. (15.8 ). 1.65. 1.11 – 2.45. Patient age. Surgeon age Time to recovery. Economic analysis (Paper IV) The costs for the different items used in this cost-minimisation analysis are listed in Table 10. For the TEP group, the index operation was € 710 (36%) more expensive for the hospital. When including community costs in order to calculate the total, this difference decreased to € 292 (7%) (Table 11). For the small group of patients that had a recurrence or complication, the total cost over five years was € 9559 for the TEP and € 8018 for the Lichtenstein repair.. Table 10. Cost in € per item. Anaesthesia time (per minute). 13.0. Surgeon (operating time) (per minute). 1.3. Recovery room (per minute). 1.0. Capital cost for laparoscopic equipment. 51.9. Non-reusable laparoscopic equipment Mesh, TEP respectively Lichtenstein. 412 21 resp. 11. Hospital care (day). 378. Sick leave (day). 179. 40.

(201) Table 11. Costs in € per patient for index operation, recurrences and complications. P. TEP. Lichtenstein. Diff. in costs. (cost). (cost). (TEP-Lichtenstein). 2662.7 107.9 2770.6. 1952.1 23.4 1975.5. 710.6 84.5 795.1. <0.001 <0.001 <0.001. 1670.6. 2250.2. -579.6. 0.001. 96.4. 19.9. 76.5. 0.013. 1767.0 4537.6. 2270.1 4245.6. -503.1 292.0. <0.001 0.024. Hospital cost Index operation - Recurrences & complications - Total hospital cost. Community cost - Sick-leave (days) for index operation - Sick-leave (days) for recurrences & complications Total community cost. Total cost. 41.

(202) Discussion. General remarks The introduction of laparoscopic cholecystectomy for the treatment of gallstones was a paradigm change for abdominal surgery. This innovation was not a result of evidence-based medicine, since laparoscopic surgery was introduced before any RCTs for comparison with the gold standard - open cholecystectomy - had been carried out. So far, this has been a common way of introducing new surgical methods or technical devices. But even if a new surgical method seems promising, it is of the utmost importance to evaluate it scientifically. When inguinal hernia repair by laparoscopy emerged, several studies were started in different countries51, 80, 93, 131 to compare it with the open technique. Randomised controlled trials are considered to be of very high scientific value for clinical research, especially if they are large, multicentric and well performed. Since inguinal hernia is a common disease, a large number of patients can be randomised within a reasonable time frame5. Furthermore, the surgical methods which are compared can be relatively well standardised. Some surgeons, however, may find this discouraging as they might be reluctant to abandon their well-established techniques. In the SMIL-2 trial, methods and criteria for repair and postoperative care were strictly standardised, and we believe there was strong adherence to this. Standardisation is probably more easily adhered to if a trial is performed at one centre with a limited number of surgeons rather than multicentric. On the other hand, this limits the generalisability of technique and study results. A RCT with its inclusion and exclusion criteria will restrict the group of patients that are under investigation. This will also influence the generalisability of the results. On the other hand, the advantage of a RCT is that the comparison of the methods in question is carried out in a specified setting where confounding factors are widely eliminated. In our trial, we excluded several potential confounding factors such as recurrent, bilateral and large (scrotal) hernias. This made the investigated group rather homogenous but still representative for the total population of individuals with primary inguinal hernia. Another factor of importance, which is specific for surgical trials, is the varying experience and dexterity among the participating surgeons. 42.

(203) Furthermore, it is a well known fact that the learning curve for laparoscopic hernia repair is rather long and perhaps as many as 250 operations are required to fully master the technique26, 132-135. The trial’s laparoscopic surgeons were skilled in laparoscopic surgery but less experienced with the TEP technique, with just a minimum of 25 previous procedures required for participation26. In the Lichtenstein group, all surgeons were considered to be experienced with the method. The study might thus have been biased in this respect. Additionally, the Lichtenstein repair may be easier to perform for the average surgeon than TEP. In a RCT, results are commonly accomplished by experts under ideal conditions, which is referred to as efficacy. Effectiveness, on the other hand, refers to results obtained in routine practice by generalists136. Thus, the present trial may be regarded as an in-between efficacy and effectiveness evaluation. It is mandatory to have a data safety monitoring board for the protection of participating patients when performing a RCT. When this trial was planned, however, this was uncommon and not demanded by the local ethics committees. As a consequence, the trial became somewhat overpowered. Poorly performing surgeons could theoretically have been identified at an earlier stage and, thus, some recurrences might have been avoided. The laparoscopic group had more ‘dropouts’, and we believe there were two main reasons for this. First, the surgeons performing the laparoscopic procedure were more concerned about previous lower abdominal surgery and hernia extension into the scrotum. The surgeons carrying out the open procedure could more easily overlook these factors. Secondly, the anaesthetists were more reluctant to administer general anaesthesia to patients with a high ASA grading for an elective laparoscopic operation. This merely reflects the limitations of laparoscopic hernia repair at that time. In this trial, participating surgeons did not perform both methods. This might introduce a selection bias affecting the external validity. To do otherwise, however, could be regarded as a confounder since the Lichtenstein surgeons were not familiar with the TEP technique and the aim of this trial was a strict comparison of the two methods in question. In one aspect the trial failed, namely regarding the lack of registration and followup of patients who declined participation. A contributing factor to this was the multicentric design with a large number of participating surgeons. A selection bias can therefore not be excluded.. Short-term results (Paper I) The short-term results of this trial clearly favour the TEP technique, with less postoperative pain and analgesic consumption, shorter sick leave and shorter time to full recovery. 43.

No results found