Laparoscopic versus open inguinal prospective trial
benefits patients because it reduces pain and enables earlier mobilisation. There is concern that laparoscopic hernia repair may enter surgical practice without proper evaluation. We have done a randomised, prospective study comparing laparoscopic and open inguinal hernia repair performed under day-case general anaesthesia. 150 patients were randomised to have laparoscopic (group L) or open (group O) herniorrhaphy. Group L underwent transabdominal stapling of preperitoneal Prolene mesh. Group 0 underwent open repair, with a tension-free nylon darn. Postoperatively patients completed pain analogue scales eight times over 7 days, and use of analgesia was recorded. Time of return to normal domestic activity and to work was assessed. The groups were similar in age, sex, and body surface area. Self-administered co-proxamol was a median of 18 tablets (1 tablet=325 mg) in group O (n = 75) and 6 in group L (n 75, p < 0·001). Overall mean pain analogue score was 3·1 (SD1·8, n = 70) in group O and 1·8 (SD 1·1, n=71) Return to normal domestic activity in group L (p<0·0001). 7 in was a median of days group O (n= 72) and 3 days in group Return to work was a median of 28 L (n=73) (p<0·001). O in and 14 days in group L (n=40) (n=39) days group
in particular minimally invasive surgery, has been described as a threat to health care.! One concern is that laparoscopic hernia repair2,3 may enter surgical practice without proper evaluation. The Royal College of Surgeons of England has recommended4 that appropriate studies to assess the new technique should be done. To our knowledge there have been no randomised prospective trials comparing a standard operative hernia repair with the laparoscopic procedure. In this study we have assessed postoperative pain, the incidence of complications, and the time taken to return to normal activity after the two types of repair.
(p<0·002). These data suggest that laparoscopic hernia repair induces pain than open hernia repair, and enables patients to return to normal activity and work more quickly. less
Lancet 1994; 343: 1243-45
Patients and methods Between June and December, 1992, laparoscopic hernia repair was done in 35 patients as a feasibility study and to optimise the technique. From January, 1993, all patients aged between 18 and 85 years referred for elective inguinal hernia repair to two surgical teams at Whipps Cross Hospital, London, were selected to have either laparoscopic (group L) or open (group 0) hernia repair. Patients were randomised to these groups with a blind envelope system; the seal was broken in the anaesthetic room before surgery. Written informed consent was obtained from each patient. Patients in whom pneumoperitoneum could not be established were excluded as were those who were unfit for general anaesthesia, were pregnant, or who had irreducible hernia, systemic or local infection, or psychiatric conditions precluding consent. The study was approved by the district ethics committee.
Operative procedures All procedures were performed under general anaesthesia by a consultant surgeon (JMW) or a senior registrar (DLS). Premedication of diclofenac (100 mg) was administered per rectum, one hour before surgery.
Open hernia repair The skin and external oblique aponeurosis opened and the hernial sac dissected out from surrounding structures. The sac was opened, and any contents reduced. Indirect sacs were then ligated, divided, and excised. In 30 patients with direct hernias, the posterior wall of the inguinal canal was plicated with braided nylon. In all patients, repair was undertaken with a tension-free interlocking nylon darn between the conjoint tendon and the inguinal ligament. The external oblique aponeurosis was closed with chromic catgut and skin with interrupted prolene. Laparoscopic hernia repair A pneumoperitoneum was established (Veress needle with CO2 insufflation). A 10 mm port was inserted through the umbilicus to carry the laparoscope, which was connected via a television camera to video monitoring equipment. Two further ports were inserted to carry the surgical instruments-a 10/12 mm port (Ethicon, Edinburgh) to the right of were
the umbilicus and a 5 Professorial Surgical Unit, St Bartholomew’s Hospital, London EC1A 7BE, UK (D L Stoker FRCS); MRC Biostatistics Unit, University Forvie Site, Cambridge (D J Spiegelhalter PhD); and Department of Surgery, Whipps Cross Hospital, Leytonstone, London, UK (R Singh MB, J M Wellwood FRCS)
Correspondence to: Mr J M Wellwood
port on the left. The hernial orifices and inspected, and the hernial sac was retracted into the peritoneal cavity. The sac and surrounding peritoneum were dissected away from the anterior abdominal wall and reflected posteriorly. Careful attention was paid to the inferior epigastric vessels, external iliac vessels, and the spermatic cord. A 12 x8 cm Prolene mesh was placed over the defect, and stapled
Comparison of patient data and hernia types
and work. These
median, and interquartile range within randomised groups. Rather than individual
significance tests each time a pain score was single comparison based on mean score over eight measurements was made. Statistical significance of between-group comparisons was assessed by X} tests for discrete variables, non-parametric U tests for continuous variables, and log-rank tests for times to return to normal activity and work. recorded,
Time after surgery
(EMS multifeed staplegun, Ethicon) along its upper border and into the pectineal ligament above the pubic ramus. To safeguard the iliac vessels and nerves, no staples were placed below the level of the inguinal ligament. The peritoneum was then stapled back over the mesh to exclude it from the peritoneal cavity. Ports were removed under direct vision and the skin sutured.5 For patients in both groups, 0-5% bupivacaine hydrochloride was infiltrated into the wound before closure. Patients graded ASA (American Surgical Association) 1 or 2 were discharged home the same day and all others on days 1 or 2 postoperatively if feasible. All patients were told that they could return to work as soon as they felt able to do so, and that early return to work would not cause recurrence of the hernia.
Outcome measures Three criteria
used to measure postoperative pain. (1) Inpatient use of postoperative analgesia was measured by the number of times non-steroidal anti-inflammatory medication was administered. Both diclofenac (intramuscular) and co-proxamol (oral) were used on the ward. (2) Outpatient use of analgesia. All patients were given 20 co-proxamol tablets (11 tablet = 325 mg) to take home and were instructed to use them as required for pain relief. Unused tablets were returned and counted after one week. (3) At eight specified postoperative times, as shown in figure 1, patients recorded their degree of pain by marking an unlabelled 10 cm visual analogue scale, which was then expressed as a score between 0 and 10. Time of operation in minutes and total inpatient time in hours were recorded. Postoperatively, patients were followed up in clinic at one and six weeks, and every three months thereafter. Standard questions were asked to assess the interval from operation to return to normal activity and, for patients in employment, the number of days off work. were
Results From January to November, 1993, 150 patients (75 in each group) took part in the trial, with all patients being treated according to random allocation. One open procedure for hernia failed because of the presence of a previously implanted steel mesh and was converted to a laparoscopic repair. Data for this patient were included with group O. The ratio of open to laparoscopic procedures was similar for each surgeon (JMW 37 to 38, DLS 38 to 37). Median hospital stay was 10 hours in both groups, with 83% of patients in each group leaving hospital on the day of recurrent
The two groups were broadly similar with respect to baseline measures (table 1). The laparoscopic operation took longer, with more variability in operating time, than did the open procedure. For unilateral repair operating time was 35 min (interquartile range 30-40, n 66) for the open procedure and 50 min (37-70, 67) for the laparoscopic procedure (p < 0-001). The times for bilateral repair were 60 min (56-62, n=9) and 92 min (51-114, n=8) (p=0-41), respectively. There was, and continues to be, a downward =
Statistics The trial protocol specified a sample size of 140. The pilot study indicated that a 5000 reduction in use of oral analgesics after the laparoscopic procedure was both feasible and important. This sample size provided a power of 80% to detect such a difference at the 5% level and enabled the study to be completed within a year. Baseline comparisons of randomised groups were carried out with respect to age, sex, body surface area, ASA score, and type and size of hernia. Outcome measures were pain scores, use of oral analgesia, number of hours in hospital, and time to return to normal
2: Use of self-administered
patients. The complication rate was 21 % in group Oand8% in group L(= 4-31, 1 df, p < 0005, table 2). There have been no recurrences in either group to date, with a mean follow-up of seven months. Discussion
Days after surgery 3: Time to return to normal
not back to normal
activity. Bottom: proportion still
back to work.
operating times as experience is gained with the new technique. Median operation time for a unilateral laparoscopic procedure was 60 min for the first 20 cases and 30 min for the last 20 cases in the trial. Narcotic analgesia was not required in either group. Inpatient administration of non-steroidal analgesia amounted to a mean of 0-7 episodes in group 0 and 05 episodes in group L. The difference was not significant. 9 patients did not complete analogue scales correctly at home. Mean pain analogue scores at each of the eight measurements were significantly lower in group L than in group 0 (figure 1) because there were more patients without pain and fewer patients with severe pain in group L. Overall mean pain score was higher in group 0 than in groupL(3 [SD 18],n=70vs 1.8 [1’1], n = 71; P < 0’001). Group L patients also took significantly fewer oral coproxamol tablets than did group 0 patients (figure 2). Times to return to normal activity and work are shown in figure 3.5 patients did not provide data on return to normal activity (group 0, n = 72; group L, n = 73). 40 patients in group L and 39 in group 0 were in employment. Group L patients returned to work significantly earlier than did
That laparoscopic hernia repair involves three small skin incisions (unilateral and bilateral repairs) with no division of muscle or aponeurosis and no cord mobilisation almost certainly accounts for the findings of reduced pain and early mobility in our study. The theoretical disadvantages of the laparoscopic approach are potential difficulties associated with induction of a pneumoperitoneum, and the risk of trauma to structures such as the spermatic cord and iliac vessels.6 The technique is safe in the hands of experienced laparoscopic surgeons, and in our series morbidity was higher in patients having open repairs (table 2). A proportion of patients in both groups expected to be immobile and off work for longer than necessary, and this notion was in some cases encouraged by their general practitioners. To obtain maximum benefit from the laparoscopic repair, it will be necessary to re-educate both patient and general practitioner about early mobilisation.7 Both types of repair can be done on a day-care basis, but the laparoscopic operation is more expensive-mainly because of the use of disposable stapling devices. The extra materials used were costed at [168. The advantage of the laparoscopic technique is therefore not to the finances of the hospital, but to the community, with fewer days lost from work. Since there are more than 80 000 inguinal hernia repairs in the UK every year,8 the laparoscopic procedure would represent substantial economic saving in lost work
days. The short
postoperative follow-up of our patients of the important issue of late hernia precludes recurrence. However, it is likely that when the mesh is initially well stapled, fibrosis will occur through the assessment
interstices of the mesh and recurrence will prove to be less common than with open repair. We conclude that the new technique of laparoscopic inguinal hernia repair causes less pain than the conventional operation and enables the patient to return to work and normal activity more quickly. Thus, it represents a significant advance in the management of a common surgical condition. acknowledge the invaluable assistance of the theatre and day-care nursing and administrative staff, without whom this trial would not have been possible. We also thank Mr M Pietroni, Mr P Frecker, Dr E Clements, the surgical house staff, and in particular Dr Peter Hutchinson and Dr Cameron Hattrick, the senior surgical We
housemen, for their assistance in theatre. References 1
Macintyre IMC. Laparoscopic herniorrhaphy. Br J Surg 1992; 79:
1123-24. Rutkow IM.
4 5 6 7 8
Table 2: Complications
Surgical innovation under scrutiny. Lancet 1993; 342:
Laparoscopic hernia repair: the socioeconomic tyranny of surgical technology. Arch Surg 1992; 127: 1271. Clinical guidelines on the management of groin hernia in adults. London: Royal College of Surgeons of England, July 1993. Corbitt JD. Transabdominal preperitonieal herniorrhaphy. Surg Laparosc Endosc 1993; 3: 328-32. McMahon AJ, Baxter JN, O’Dwyer PJ. Preventing complications of laparoscopy. Br J Surg 1993; 80: 1593. Stoker DL, Wellwood JM. Return to work after inguinal hernia repair. Br J Surg 1993; 80: 1354-55. Corson RJ. Management of groin hernias in adults. Ann R Coll Surg Engl 1993; 75 (suppl): 125-28.