Abstract

INTRODUCTION

Cholelithiasis is the most common disease of the biliary tract, and it is estimated that nearly up to 10%–15% of the adult population in developed countries have gallstones.[1,2,3] Laparoscopic cholecystectomy (LC) is the ‘gold standard’ operation for symptomatic gallstones, but despite great success with the operation, there still remain some patients who continue to have symptoms after surgery.[4,5,6] The presence of symptoms after cholecystectomy is termed ‘post-cholecystectomy syndrome’ (PCS), and includes symptoms such as nausea, vomiting, gas bloating, jaundice, diarrhoea or abdominal pain occurring at any time after cholecystectomy.[4] Certainly, there is no clear-cut definition of PCS, and even disorders unrelated to the biliary tract are included if they cause symptoms after cholecystectomy;[7,8,9] given this, one can appreciate the difficulty in assigning an exact figure for its incidence. A highly variable incidence of 5%–63%[9,10] has been reported by various authors; although some feel that a 10%–15% incidence of PCS is reasonable, no reasons have been proposed for this.[11]

PCS is distressing to a patient who, even after having undergone a ‘successful cholecystectomy’, needs to be investigated thoroughly once again to look for any cause of symptoms which may be related to (i) the diagnosis (overlooked stones or neoplasm), (ii) operative procedure (stricture, cystic duct remnant, adhesion or injury to common bile [CBD]), (iii) functional disorders affecting the sphincter of Oddi or (iv) diseases of the surrounding organs.[12] The common presenting symptoms are dyspepsia and pain,[8,9,13] but there is very little in the history to suggest anything about their origin.[5] PCS can occur early (occurring in the post-operative period) or late (after months or weeks) and includes a large number of disorders; both, biliary and non-biliary diseases in almost equal incidence.[6,9] The most common reported biliary cause of PCS is ductal calculi, and the most common reported non-biliary cause is gastric/duodenal ulcer.[11,12] Early and severe symptoms are usually considered to represent a complication of cholecystectomy.[9]

Any patient presenting with PCS needs symptomatic treatment, and a detailed workup to diagnose the exact cause of symptoms; liver functions and abdominal ultrasound are usually the first line of tests that are ordered, and further investigations tailored accordingly.[9]

Despite a large volume of literature on PCS, surprisingly, we could come across very few articles on this topic from our country despite LC being one of the most common operations being performed here.[14,15,16] In view of this, the present study was undertaken to assess the incidence and causes of PCS in our set-up and if possible, provide an insight into this problem.

MATERIALS AND METHODS

After Ethical Committee approval, data were prospectively maintained for 207 patients who underwent elective LC in our department, with the objective of assessing the incidence of post-cholecystectomy symptoms, their cause (if any) and the risk factors for developing symptoms (by analysis against patient particulars, symptoms and pre-operative investigations). Informed consent to participate in the study was taken, and data were maintained for all patients with symptomatic gallstones (without choledocholithiasis, gallbladder cancer or diagnosed peptic ulcer disease) who agreed to participate in the study.

Standard four-port LC was performed under general anaesthesia in all patients. The details of surgery were documented; and once discharged, these patients were followed up for 6 months. On follow-up, symptoms, if any, were documented and investigated only if they persisted beyond 1 month of surgery. Liver function tests and abdominal ultrasound were the first line of investigation, and further investigations such as endoscopic retrograde cholangiopancreatography (ERCP) were asked for only if indicated.

Symptoms were recorded on post-operative follow-up, and analysed each time against various parameters such as patient’s age, sex, BMI, pre-operative duration of symptoms, pre-operative emergency admissions (for acute cholecystitis), presence of comorbidities, previous abdominal surgery, pre-operatively elevated alkaline phosphatase, pre-operative ultrasound findings (number of calculi and CBD size), type of surgery performed (successful LC or conversion) and duration of surgery.

Data collected from the study were analysed using Chi-square test, Student’s t-test and Mann-Whitney U test. The value of significance was taken as P < 0.05.

RESULTS

Of the total of 207 patients, there were 22 male and 185 female patients, giving a male-to-female ratio of 1:8.5. The average age was 44.4 years (range: 12–79 years) with a slightly higher mean age in males (45.77 years) as compared to females (44.23 years). Overall, the duration of symptoms before surgery ranged from 1 month to 45 months; only 16 (7.7%) had the previous history of hospitalisation for acute cholecystitis. All LCs were performed electively (after 6 weeks of any acute episode) as is our protocol. Forty-one patients (19.8%) had co-morbid diseases (diabetes mellitus, hypertension and hypothyroidism); 37 (17.8%) had undergone the previous non-biliary abdominal surgery.

LC was successfully performed in 189 patients (91.31%) and converted to open in 18 (8.69%). The most common cause of conversion was unclear anatomy in the Calot’s triangle; no bile duct injuries occurred in the period of study. The duration of surgery ranged from 30 to 210 min (mean: 66.45 min) and the duration of post-operative stay ranged from 1 to 6 days (mean: 1.32 days) [Table 1]. The patients were followed up at 7 days, 1 month, 3 months and 6 months post-operatively, and symptoms, if any, were documented. Patients who remained symptomatic at the end of the 1^st month were investigated with liver functions and abdominal ultrasound; patients who had an elevated alkaline phosphatase along with a dilated CBD on ultrasound were taken up for ERCP.

Table 1

Outcome of laparoscopic cholecystectomy (present series)

On the first follow-up visit, on the 7^th post-operative day, 120 patients (58%) had complaints of dyspepsia, heaviness after meals, flatulence, nausea and pain abdomen whereas 87 (42%) were asymptomatic. The most common symptom was dyspepsia, and no statistically significant relationship was observed between the symptoms and age, sex, BMI, pre-operative duration of symptoms, previous abdominal surgery, pre-operative ultrasound findings (number of calculi and CBD size), type of surgery performed (successful LC or conversion) or duration of surgery. However, a significant relationship was seen with pre-operative emergency admission (for acute cholecystitis), the presence of co-morbidities and pre-operatively elevated alkaline phosphatase.

On the 2^nd post-operative visit at 1 month, the number of symptomatic patients dropped to 81 (39.1%). A similar relationship as seen in the previous visit was observed between the various parameters analysed, with only two differences: (i) a statistically significant relationship developed between symptoms and patients who had undergone previous abdominal surgery and (ii) pre-operatively elevated alkaline phosphatase was no longer a significant factor in the symptomatic group of patients.

At this point of time, all symptomatic patients were advised liver function tests, serum amylase, serum lipase and abdominal ultrasound. All these patients had normal amylase and lipase, but alkaline phosphatase elevation was seen in 16 (19.75%), and of these 16 patients, the CBD was dilated in seven patients. These seven patients were taken up for ERCP that revealed normal bile duct in 4, duodenal ulcer in 2 and no significant findings in 1.

The patients were followed up at 3 and 6 months post-surgery, with an almost similar picture. It was observed that the statistically significant relationship that had been seen earlier between symptoms and pre-operative emergency admission (for acute cholecystitis) and presence of comorbidities persisted, whereas all the other factors including pre-operatively elevated alkaline phosphatase (significant on 1^st visit), and the previous abdominal surgery (significant on 2^nd visit) were not significant in the affected group of patients. Only 27 patients (13.0%) remained symptomatic at 6-month follow-up [Table 2].

Table 2

Post-operative Follow up (n=207)

DISCUSSION

In World literature, the incidence of PCS varies widely, between 5% and 63%,[9,10] but despite the volume of LC being performed in India, there is very little data on the incidence, presentation and management of PCS. The earlier report of Anand et al.[15] who followed up 171 patients prospectively for nearly 3 years had an incidence of 18.13% whereas a more recent study from India reported an incidence of 27%.[16] For all intents and purposes, any patient who presented with any symptoms after LC was labelled PCS in our study, and after an initially high incidence of 58%, only about 13% remained symptomatic after 6 months. To the best of our knowledge, this fall in the incidence of symptoms after cholecystectomy has not been documented previously before, and might quite possibly reflect the physiological adjustments within the body following loss of the reservoir function of the gallbladder[4,17,18] as well as the tissue response after dissection in the Calot’s triangle.

There are conflicting reports in the literature about factors that contribute to the development of PCS.[5,6,11,12,13,18,19,20] In the present series, there was no statistically significant relationship between post-operative symptoms and age, sex and BMI; the absolute incidence is higher in females probably because more females undergo cholecystectomy (1:8.5 male-to-female ratio in the present series) but this does not translate into a ‘risk’ factor for developing symptoms post-operatively. We did not find any significant relationship of PCS with age despite symptoms being most common in the age group of 41–60 years (46.4%); nor did we find any relationship with obesity, which has is a well-documented risk factor for gallstones.[21,22] Neither the duration of pre-operative symptoms, the surgery performed (LC or conversion), nor the duration of post-operative admission had any effect on PCS in the present series, but analysis of our cohort during follow-up showed that there were two constant factors (previous attacks of acute cholecystitis and presence of comorbid illnesses) and two temporary factors (elevated pre-operative alkaline phosphatase and prior abdominal surgery) that significantly affected the development of PCS.

Prior attacks of acute cholecystitis have been found to be significantly associated with PCS,[10] and the mechanism by which acute cholecystitis leads to PCS has been proposed by the theory of referred pain from the gallbladder (which can persist for years due to central neuroplastic changes), continuing or manifesting post-operatively.[23] In fact, to reduce repeated attacks of acute cholecystitis; and therefore, reduce the incidence of PCS, it was proposed that early cholecystectomy be performed,[24] but unfortunately, we are unable to comment on this aspect since we electively operate our patients after 6 weeks of an acute attack. The other constant factor for the development of PCS in our series was the presence of co-morbid illnesses, mainly diabetes. Gastric dysmotility as well as biliary dyskinesia (as a result of autonomic dysfunction) is well documented in patients with diabetes;[25,26] the significant association between diabetes and PCS seen in our patients might be explainable on the basis of impaired gastric motility persisting or manifesting post-operatively.

We also observed two temporary factors, that is, remained significantly associated with post-operative symptoms for a short while and then reverted. These were (i) elevated pre-operative alkaline phosphatase and (ii) prior abdominal surgery. Patients with pre-operatively elevated alkaline phosphatase were taken up for surgery after a normal calibre of the common duct on ultrasound. Surprisingly, many of these patients presented with early symptoms after cholecystectomy (in the 1^st week), and this was a significant association. A few authors have suggested that early symptoms after cholecystectomy arise due to complications of the operation,[6,9] but in the present series, we did not have any overt LC related complications that could be blamed for these early symptoms. In our patients, the association of early PCS and elevated pre-operative alkaline phosphatase reverted on subsequent visits and was no longer significant for the rest of the duration of the study; we feel that the early and temporary nature of this response possibly reflects some degree of cholestasis or underlying sphincter dysfunction that is temporarily overwhelmed due to an increased flow of bile through the common duct after LC and manifests in the early post-operative period as symptoms, settling over time once the choledochus adjusts to increased bile flow through it. We hypothesise that sphincter spasm after cholecystectomy[27,28] and local tissue trauma during surgery (leading to a pericholedochal inflammatory response) might also contribute to symptoms early after cholecystectomy.

The second temporary or transient factor was previous abdominal surgery. In the present series, patients who had undergone abdominal surgery previously became symptomatic by the 2^nd visit, that is, by the 1^stmonth post-operatively. By the 3^rd month, this was no longer a significant factor for PCS, thereby indicating a possible transient ‘intestinal phase’ that becomes prominent once the biliary phase of PCS subsides. Possibly, patients with prior abdominal surgery manifest overt symptoms due to the presence of intra-abdominal adhesions, but this is our ‘unfounded’ hypothesis. Although intra-abdominal adhesions have been implicated in the causation of symptoms after cholecystectomy,[29] these reports are from the era of open cholecystectomy and probably do not reflect the situation of the present times.

The phasic response (biliary and intestinal) noted in our series has not been appreciated previously in literature; possibly, classifying symptoms into biliary (nausea, vomiting, food intolerance and tenderness), dyspeptic (flatulence, heartburn and belching)[30,31] and intestinal (abdominal cramps and diarrhoea) might help in a better understanding of PCS and its management. Although we did find some biliary symptoms appearing in the early period and intestinal symptoms later (correlating with the phasic response), dyspeptic symptoms remained predominant throughout the period of study, and persisted over a longer follow-up; this might be classified as a ‘persistent’ phase of symptoms where dyspeptic symptoms persist over a long-term period. Other authors too, have shown that dyspeptic symptoms are far more common in the post-cholecystectomy period,[5] thereby indicating that there may be many more mechanisms involved in PCS than simple alterations in the flow of bile. Unfortunately, the short-term follow-up of only 6 months in our series does not allow for any further observations to be made regarding this ‘persistent’ phase of post-cholecystectomy symptoms.

In the present series, 81 patients who remained symptomatic after 1 month of surgery were investigated, and only seven (3.38%) were taken up for ERCP after having been found to have significant dilatation of the CBD in association with elevated post-operative alkaline phosphatase. Of these seven, a definite cause could be detected on ERCP in only two (duodenal ulcer); four patients had a dilated CBD, and in one, it was normal. In the absence of manometry, it is ‘unscientific’ to label these dilated systems as sphincter of Oddi dysfunction, since this could very well be a normal physiological dilatation after cholecystectomy;[17] thus, only 2/207 patients, or 0.97% of our patients had a detectable cause for their symptoms. This fact, coupled with our findings of an ‘intestinal phase’ of PCS that settles by the 3^rd post-operative month, it might be prudent to delay investigating symptomatic patients for at least 3 months post-operatively. Magnetic resonance cholangiopancreatography (MRCP) has been shown to be a better option than ERCP,[7,9] but since MRCP was not available in-house during the period of study, we were unable to analyse this aspect further.

Going over our data once again, a pattern of three groups emerges; the first, of patients with transient symptoms; the second, of those in whom symptoms persisted up to 6 months (13%); the third, those in whom there was a definite cause of post-cholecystectomy symptoms. We feel that patients who continue to remain symptomatic on a long-term follow-up in the absence of any detectable cause are the ones who actually deserve the label of ‘PCS’, and all others should be termed either post-cholecystectomy symptoms (transient) or with what they are diagnosed on investigation (e.g., duodenal ulcer disease).

Towards the end, a few points may be worth stressing on in light of the present series; the first is that it should be kept in mind that patients who have pre-operatively elevated serum alkaline phosphatase but a normal diameter of the CBD on ultrasound do tend to become transiently symptomatic after surgery. This might help in counselling patients pre-operatively or in relieving apprehensions after surgery. The second is that it might be better to delay investigating symptomatic patients for at least 3 months after LC since most symptoms resolve by this time; however, apprehensions (surgeons as well as the patients) may prove to be a stumbling block in delaying investigations; the earlier a normal report, the better it is for all concerned! And thirdly, the term PCS should be reserved for patients who remain symptomatic beyond 6 months in the absence of any demonstrable cause. Unfortunately, the short duration of follow-up (only 6 months) in our study does not give us any further information about this group of patients.

CONCLUSION

LC is a safe procedure in trained hands, especially when the procedure is converted early and there is no insistence in persisting with the minimally invasive approach in difficult cases. More than half of patients will have some symptoms after LC; dyspeptic symptoms are the commonest and persist over a long-term; biliary symptoms are seen early (biliary phase) followed by intestinal symptoms (intestinal phase), both of which are transient. Patients with previous attacks of acute cholecystitis and those with co-morbid diseases are more prone to develop persistent symptoms. Very few patients (<1% in the present series) had a detectable cause of symptoms, and only patients with persisting symptoms and no obvious cause on investigation should be labelled as PCS. However, there seems to be a large lacuna in the understanding of post-cholecystectomy symptoms, especially from India, and there is a need for larger series or multicentric trials to understand in a better way, the physiology of altered bile flow and symptoms after LC and the cause of persistent, long-term symptoms in such patients.

REFERENCES