Sir
We disagree with the conclusions drawn in the paper by Clavien et al. (Br J Surg 1990; 77: 73742). The paper quite rightly points out that we are dealing with an elderly group of patients many of whom have serious concomitant disease. These patients are often markedly dehydrated and septic at the time of operation due to the unfortunate but common occurrence of delayed presentation and diagnosis. Yet the conclusions are that we should (when feasible and when cholelithiasis has been shown) carry out an extended and sometimes difficult operative procedure. Clavien et al. base their claims on the following points.
First, that there appeared to be no significant difference in mortality between the three groups of patients. If, however, we look at the patients who had surgery focussed on the obstructing stones alone it appears that this is not an homogeneous group. Two had small bowel resections and one a Hartmann's procedure. These patients surely form a separate group (complicated gallstone ileus) as opposed to enterolithotomy alone to which we feel the one stage procedure should be compared. The authors unfortunately d o not make it clear which operative procedures preceded individual deaths though it would appear that two of the seven deaths occurred after relaparotomy for leakage from intestinal suture lines. If these represent the patients who had bowel resections then this could possibly influence the results depicted and show the relatively simple procedure of enterolithotomy alone to have a lower mortality.
Second, they argue that removing the gallbladder prevents further symptoms such as; ( I ) repeated gallstone ileus, (2) cholangitis and (3) carcinoma of the gallbladder, and while this is an undeniable fact, it raises the question: 'Is the risk of developing these complications greater than the risk of removing the gallbladder at the initial operation?' This question remains unanswered by the paper.
Recurrent gallstone ileus in most reports has a lower incidence than in this review. The two early recurrences may even be explained by failure to detect a second stone already within the bowel at the original operation. Even if this were not the case it is likely that 80 per cent of stones passing through afistula will pass the bowel without problem'.
Acute cholangitis following gallstone ileus should only occur in circumstances when the fistula or cystic duct is no longer patent (even if the organisms originate in the fistula), and there is obstruction within the common bile duct. The frequency of cholangitis with concomitant common bile duct stones reported by Safaie-Shiazi et al.* in 47 patients with biliary enteric fistula without intestinal obstruction would tend to support this view. Under these circumstances, endoscopic sphincterotomy would be the first treatment of choice in such patients, operation being reserved for those in whom this was unsuccessful.
we can find no comparison between the incidence of gallbladder cancer in patients undergoing cholecystectomy and those with biliary enteric fistulae. Even if there were a higher incidence of carcinoma of the gallbladder the carcinoma could have an aetiological role in the formation of the fistula.
It should be noted that none of the patients followed up in Clavien et al.'s study died of causes related to further biliary disease. We have based our view of this condition on our own review of 43 patients (unpublished data ).