Symptoms attributable to gallstone disease result in 60,000 cholecystectomiesper annum in the UK.1 Improvements in surgical techniques and advances in post-operative care have resulted in just over 50% of cholecystectomies being performed as day-cases, a rise from 16% in 2008–09.2, 3 The majority of patients recover uneventfully. 10% of patients may develop post-cholecystectomy syndrome (PCS) weeks to month’s later.4
Post cholecystectomy syndrome was first described in 1947 by Womack and Crider.5 It refers to the persistence of gastrointestinal symptoms following cholecystectomy and may occur in 5–47% of patients.6, 7, 8 The aetiology can be broadly divided into biliary, extra-biliary, organic and functional.9
It has been hypothesised that the majority of patients who develop post cholecystectomy syndrome may actually suffer from extra-biliary or organic disorders, such as gastroesophageal reflux disease or acute and chronic pancreatitis.10 The difficulty lies in establishing whether these disorders were present before cholecystectomy and were subsequently exacerbated by the post-operative changes in biliary kinetics, or whether they are new symptoms secondary to the procedure. Inappropriate operations may also predispose to the development of PCS.
Retained stones, de novo ductal stone formation, strictures and sphincter of Oddi dysfunction (SOD) may all present as PCS and can all be attributed to biliary aetiology post-surgery.11 The Rome III committee concluded that PCS pain could be categorized by character and location of symptoms,12 and thus assist characterisation.
Due to the different causes of PCS there are a number of management options available. These can be categorized into medical, endoscopic and surgical. The management of choice is largely dependent on the aetiology and premorbid state of the patient. Although guidelines exist for the management of the underlying cause of post cholecystectomy syndrome e.g. retained or de novo stone, there are no published guidelines on how to investigate post cholecystectomy syndrome of unknown cause for the underlying aetiology.
This systematic review aims to describe the causative aetiologies of post-cholecystectomy syndrome. Their incidence, therapeutic interventions, patient outcomes, resolution of symptoms and adverse are also described.
Materials and methods
This systematic review was registered with the PROSPERO database of systematic reviews (CRD42016035916) and reported in accordance with the PRISMA guidelines.53
A systematic search of Medline, Embase and Cochrane databases was conducted. The search strategy comprised the keywords and MeSH terms: ‘Post Cholecystectomy Syndrome’, ‘aetiology’ and ‘management’ for the time period 1990–2016. Search strategies for each database are included in the supplement. The inclusion criteria were as follows; studies reporting 10 cases or more, reporting PCS in adults (defined as 18 years and over) following laparoscopic cholecystectomy, published in English from 1990 onwards. A hand search of bibliographies was conducted to identify any additional articles.
Open surgery operations were excluded to reduce heterogeneity of the included studies. The first laparoscopic cholecystectomy was performed by Professor Müher in 1986.13 Studies of laparoscopic cholecystectomy published prior to 1990 were therefore excluded, as outcomes in these preliminary studies were likely to be biased by operator experience.
All abstracts were independently reviewed by two of the authors. Any disputed abstracts were further discussed and put to the senior author for arbitration. The full texts of eligible abstracts were reviewed and screened for study eligibility independently by two authors. All study types; randomised control trials (RCTs), case–control, cohort and case series were eligible for inclusion. Results from case series were excluded from any pooled analysis relating to intervention, as they carry high risk of bias.49 Case series data on aetiology were included on a descriptive basis. Where there was insufficient data to make a decision about eligibility study authors were contacted for additional information.
Two authors independently extracted the following data from each study:
Aetiology – defined as the named cause of symptoms as described in each paper, reported for each participant
Symptoms – defined as the reported nature of the symptom, timing of symptoms reported against time from initial cholecystectomy
Treatment – frequency of medical, endoscopic or surgical intervention
Primary outcomes were the number of participants with a described aetiology and the type and timing of recurrent symptoms after cholecystectomy. Secondary outcomes included the number and type of complications related to the intervention, number of participants reporting recurrent symptoms following intervention, number of attendances at primary or secondary care services after intervention and all-cause mortality.
Studies that reported a case–control or cohort study design without methodological concordance were discussed with the senior author and analysed as case series if appropriate.
Results of randomised controlled trials were converted to risk ratios (RR) plus 95% confidence intervals (CI) for dichotomous data and mean difference (MD) plus 95% CI for continuous data using Revman (Cochrane Collaboration, Denmark). Results of cohort and case control studies were converted to odds ratios (OR) plus 95% CI for dichotomous data and mean difference (MD) plus 95% CI for continuous data and pooled using random effects meta-analysis if appropriate. Where studies reported standard error, this was converted to standard deviation.
Risk of bias
Risk of bias in RCTs was assessed using the Cochrane risk of bias tool14
Case control and cohort studies were assessed using the Newcastle–Ottawa Scale15
The search identified 1077 articles (Fig. 1). Duplicate articles were removed, 104, and the remaining titles and abstracts were reviewed. 876 articles were excluded following title and abstract screen and 97 potentially eligible references were identified for full paper review. On full text review, 69 were excluded and we were unable to obtain the full paper for 7 references by inter library loan despite a national search, leaving 21 articles for data extraction (15 case series, 2 cohort studies, 1 case control and 3 RCTs).
All of the studies recruited participants from a single centre. They were published from a range of countries; three from India18, 24, 26 and USA10, 28, 32 respectively, two from the UK23, 34 and one from China,33 Thailand,31Austria,29 Hungary,27 Romania,25 Ireland,22 Denmark,6 Australia,21 Italy,20France,19 Norway,17 Sweden30 and Germany,16 providing good geographical diversity. The studies were conducted between 1990 and 2014. The study characteristics are shown in Table 1.
|Study [year]||Country||Study methods||Total participants||Participant eligibility criteria||M:F||Age ± ge (range), years||Duration of study||Intervention(s)|
|Bouzo 199016||Germany||RCT (cross-over)||20||Colic-like pain, meteorism, flatulence, dyspepsia||6:14||52.4||8 weeks||Cisapride versus placebo|
|Farup 199117||Norway||RCT (cross-over)||19||Recurrent abdominal pain or discomfort in the right upper quadrant of the abdomen in relation to meals of ≥3 months duration||0:19||56 (44–71)||10 weeks||Cisapride versus placebo|
|Khuroo 199218||India||RCT (cross-over)||28||≥3 episodes of pain resembling biliary pain in preceding 3 months, elevated serum ALP at least twice during documented pain & sphincter of Oddi findings suggestive of spasm||3:25||35 ± 1.5||26 weeks||Nifedipine versus placebo|
|Druart-Blazy 200519||France||Retrospective cohort||37||Pain, abnormal liver biochemistry or ultrasound abnormalities and suspected Sphincter of Oddi dysfunction||7:30||56||mean 3.5 years||Opiates prior to onset of pain versus non opiates|
|Cicala 200220||Italy||Prospective Cohort||140||Sphincter of Oddi dysfunction (defined using Rome Criteria)||12:18||50 (23–66)||10–13 months||Endoscopic Sphincterotomy versus conservative|
|Bennett 200921||Australia||Retrospective Case control||72||Biliary-like pain consistent with Sphincter of Oddi dysfunction (Rome Criteria)||8:64||45 ± 12||NR||None|
|Bisgaard 20056||Denmark||Prospective Case series||150||Biliary symptoms resembling pre-operative symptoms||21:129||41 (20–79)||2 years||None|
|Caldwell 199522||Ireland||Prospective Case series||22||All cholecystectomy patients||3:19||42 (24–85)||6 months||None|
|Carlson 199223||UK||Retrospective Case series||384||Abdominal pain||24:126||54.8 (no detail)||NR||None|
|Dilawari 199024||India||Presumed Retrospective Case series||122||Post cholecystectomy syndrome (no detail)||34:88||46.5 (18–80)||NR||None|
|Filip 200825||Romania||Presumed Prospective Case series||80||Abdominal pain, rigors, jaundice or dyspepsia||32:48||56.2 ± 6.3 (35–80)||NR||None|
|Kochhar 199326||India||Presumed Retrospective Case series||18||Epigastric or right hypochondrium pain, flatulent dyspepsia||4:14||(18–61)||NR||None|
|Madacsy 200627||Hungary||Prospective Case series||85||Suspected Sphincter of Oddi dysfunction (Geenen classification)52||NR||NR||12 months||Endoscopic sphincterotomy|
|Philips 201410||USA||Retrospective Case series||12||Remnant cystic duct lithiasis||5:7||38.3 (24–57)||NR||ERCP versus Surgery|
|Quallich 200228||USA||Prospective Case series||83||Suspected Sphincter of Oddi dysfunction (no detail)||10:50||NR||NR||None|
|Rogy 199129||Austria||Retrospective Case series||322||Re-operation after cholecystectomy (no detail)||95:227||61||NR||None|
|Rolny 199330||Sweden||Presumed Retrospective Case series||17||Group I SOD (Geenen classification)52||2:15||62 (37–80)||Mean 28 months||Endoscopic sphincterotomy versus Surgical sphincterotomy|
|Soontrapornchai 199731||Thailand||Presumed Retrospective Case series||27||Recurrent pre-operative symptoms||NR||NR||NR||antispasmodics, laxatives, cisapride, antacids|
|Topazian 200432||USA||Retrospective Case series||74||Chronic abdominal pain at least 1 year after cholecystectomy||7:67||46 (22–79)||Mean 36 months||Endoscopic sphincterotomy|
|Zhou 200333||China||Case series||386||History of PCS (no detail)||155:231||(19–85)||NR||Endoscopic sphincterotomy versus balloon dilatation|
|Fullarton 199234||UK||Case series||40||Case: Group II SOD (Geenen classification)52
Control: CBD stones or no pancreaticobiliary disease
|8: 32||55.5||mean 24 months||Endoscopic sphincterotomy|
NR; not reported.
Thirteen studies described an intervention or treatment; five studies (three RCTs,16, 17, 18 one retrospective cohort19 and 1 retrospective case series31) described medical treatments; two RCT’s compared cisapride versus placebo,16, 17 one RCT compared nifedipine versus placebo,18 one cohort study compared opiate use versus non-opiate use prior to onset of pain,19 and one case control compared combinations of anti-spasmodics, laxatives, antacids and cisapride.31 Seven studies (one prospective cohort20 and six case series10, 27, 30, 32, 33, 34) described endoscopic or surgical intervention. One prospective cohort20 and five case series27, 30, 32, 33, 34 included endoscopic sphincterotomy as an intervention. Endoscopic sphincterotomy was compared with balloon dilation or surgical sphincterotomy in two of these case studies.30, 33 One case series compared ERCP with surgical intervention27 (open bile duct exploration).
All studies included male and female participants, although most included a higher ratio of females. Only one study recorded BMI.10 Some studies reported only patient weights, without a clear indication of habitus, therefore limiting inter-participant comparison. The ages of study participants ranged from 41 to 62 years.
The method of patient identification for inclusion varied considerably but mostly consisted of self-report of symptoms to the research team during clinical consultations in the prospective studies, or review of notes in the retrospective studies. The criteria for inclusion as a case of PCS differed between studies. Seven studies included participants if they had features consistent with SOD, Cicala et al.20 and Bennett et al.21 using the Rome Criteria, Fullarton and Murray,34 Madacsy et al.27 and Rolny et al.30 using the Geenen Classification. Four studies reported no detail on the criteria that constituted PCS.22, 24, 29, 33 All studies varied in the timeframe in which symptoms were captured, both in regard to the duration of time since laparoscopic cholecystectomy and length of participant follow up. Although pain characteristics were described in most studies, all failed to define how they determined if the pain was a continuation from the pre-operative period, exacerbated following a period of improved symptoms or of new onset.
One RCT,18 one cohort19 and 12 case series reported the cause of PCS in their study population (Table 2). In the first three years after cholecystectomy, gastric pathologies including peptic ulcer disease, hiatus hernia and gastro-oesophageal reflux disease were the most common causes of PCS, found in 11–100% cases. In the longer-term (more than three years post-cholecystectomy) retained ductal stones within the biliary tree were the most common cause, reported in 4–40% participants. Sphincter of Oddi dysfunction (SOD) accounted for 1.8–31% of cases in an unselected population with PCS.18, 25, 32, 33 In population pre-selected with likely SOD19, 21, 27, 28only 25–47% cases had true SOD confirmed on liver biochemistry, ERCP or manometry as per the Rome or Milwaukee Criteria.35, 36, 37 In these studies 4.1–50% cases had no cause of their symptoms identified.19, 21, 27, 28 The number of reported aetiological diagnosis did not necessarily correspond with the number of patients recruited to the study, the discrepancy has been reported as unknown in Table 2.
|Study||Total Patients diagnosis recorded for||Time since operation (years)||Abdominal – extra biliary||Biliary–organic||Biliary–functional||Post-operative||Othera||Unknown|
|Peptic ulcer disease, GORD or HH (%)||Pancreatitis
|Cystic duct remnant
|Insufficient cholecystectomy N (%)|
|Caldwell22||22||0.5||7 (100)||excluded||excluded||excluded||excluded||excluded||excluded||excluded||excluded||excluded||3 (18)|
|Khuroo18||153||>0.5||excluded||3||excluded||32 (21)||5 (3)||30 (20)||5||48 (31)||excluded||excluded||20 (13)||15 (10)|
|Soontrapornchai31||22||1–2||3 (11)||excluded||1 (4)||0||excluded||excluded||excluded||excluded||excluded||excluded||3||15 ()|
|Kochhar26||18||0–3||10 (56)||excluded||excluded||2 (11)||excluded||excluded||excluded||excluded||excluded||excluded||excluded||6 (33)|
|Topazian32||74||Mean 3||2 (5)||excluded||excluded||3 (4)||excluded||excluded||excluded||39 (27)||excluded||excluded||14 (31)||24 (43.5)|
|Filip25||80||0–5||excluded||6 (7.5)||excluded||24 (30)||8 (10)||excluded||excluded||7 (9)||3 (4)||2 (2.5)||30 (38)||0|
|Rogy30||35||Median 9||excluded||3 (9)||1 (3)||14 (40)||excluded||excluded||6 (11)||excluded||4 (11)||7 (20)|
|Dilawari24||105||0–20||excluded||5 (5)||excluded||30 (29)||31 (30)||excluded||excluded||excluded||excluded||excluded||5 (5)||34 (32)|
|Zhou33||341||unknown||6 (1.6)||excluded||excluded||243 (65)||61 (16.6)||excluded||5 (1.3)||7 (1.8)||4 (1)||Excluded||15 (4.0)||30 (8)|
|Druart-Blazy19||147||0.25–32||excluded||excluded||excluded||44 (30)||14 (9.5)||excluded||excluded||37 (25)||28 (19)b||11 (7.4)||6 (4.1)|
|Bennett21||72||unknown||excluded||excluded||excluded||excluded||excluded||excluded||18 (25)||33 (46)||excluded||excluded||excluded||21 (29)|
|Madacsy27||85||unknown||excluded||excluded||excluded||15 (18)||excluded||excluded||excluded||40 (47)||excluded||excluded||excluded||30 (35)|
|Quallich28||60||unknown||excluded||excluded||excluded||3 (5)||excluded||excluded||excluded||25 (45)||excluded||excluded||excluded||30 (50)|
Other diagnoses included biliary acariasis, chronic wound breakdown, cystitis, musculoskeletal pain, adhesions, colitis, haematological, coeliac sprue, hilar lymphadenopathy, pre-ampullary diverticulum, liver and spleen congestion from thalassaemia, Campylobacter jejuni.
Depicts post-operative causes not detailed within the text. Unknown causes account for any discrepancy noted between the total number of aetiologies listed in the paper, and the reported total number of participants for which a diagnosis was recorded for, also listed in the table.
The three RCTs that investigated medical management assessed cisapride, nifedipine or placebo.16, 17, 18 Bouzo16 randomised 20 participants to receive 10 mg cisapride, three-times daily for four weeks or placebo, Farup et al.17randomised 19 participants to receive 20 mg cisapride twice-daily or placebo and Khuroo et al.18 randomised 28 patients to 12 weeks of nifedipine at maximally tolerated dose between 5 and 20 mg or placebo.
In the cross-over RCT by Bouzo16 fewer patients experienced recurrent episodes of PCS symptoms at four weeks of treatment with cisapride (RR 0.41, 95% CI 0.22 to 0.77, 20 participants). Farup et al.17 assessed the efficacy of cisapride using patient reported pain scores and biliary drainage, but did not report discrete values for these outcomes, only median differences. The authors reported a trend towards increased symptom frequency, with mean daily pain scores increasing in the cisapride group in the latter two weeks of treatment compared with the placebo group. However this was not statistically significant. Neither study reported adverse events, or the number of unscheduled post-intervention visits to primary or secondary care.
Khuroo et al.18 reported that when receiving nifedipine, participants had a trend towards fewer episodes of pain (MD −1.6, 95% CI −3.27 to 0.07, 28 participants), lower pain scores (MD −10.1, 95% CI −14.8 to −5.39, 28 participants) and fewer visits to the emergency room (MD −1.1, 95% CI −1.94 to −0.26, 28 participants). The study did not report adverse events.
The heterogeneous nature of the RCTs precluded meta-analysis.
Druart-Blazy et al.19 examined a group of 37 patients with suspected Sphincter of Oddi dysfunction, comparing 13 participants who received regular opiates before the onset of pain with 23 who received no opiate analgesia. Participants receiving opiates had more episodes of recurrent symptoms (OR 3.15, 95% CI 0.45 to 21.95, 36 participants) over the follow-up period of 3.5 years, although the authors did not state whether patients continued their treatment regime for the duration of the study. The study did not report adverse events.
Cicala et al.20 conducted sub-group analysis of a prospective cohort study of 30 patients with a diagnosis of sphincter of Oddi dysfunction (Rome Criteria I and II). Of these patients, 22 patients were found to have a prolonged hepatic-hila-duodenal transit time and were subsequently offered sphincterotomy. 14 patients elected to undergo endoscopic sphincterotomy, and were compared to the remaining 8 patients who were managed conservatively. Significantly fewer patients that underwent sphincterotomy experienced recurrent symptoms by 13 months (OR 0.01, 95% CI 0.0 to 0.18, 22 participants). All of the patients that were managed conservatively (n = 8) experienced recurrent symptoms. This study did not report adverse outcomes or unscheduled visits to primary or secondary care post-intervention.
Risk of bias
None of the RCTs provided sufficient detail to determine the risk of bias of their methods of sequence generation or allocation concealment (Table 3). Bouzo16 and Farup et al.17 stated that they performed double-blind RCTs. Khuroo et al.18 also stated this, but when participants received the placebo drug investigators were not blinded, whereas both were blinded when participants received the study medication. This may have led to a potential source of bias. Only Khuroo et al.18 reported the use of a protocol that could be used to assess bias in selective outcome reporting. Bouzo16 were deemed to be at high risk of other bias as there was no wash-out period included in this cross-over trial. Farup et al.17 and Khuroo et al.18 both incorporated a washout period of two weeks between placebo and the study drug.
|RCT||Sequence generation||Allocation concealment||Blinding of participants and personnel||Blinding of outcome assessors||Incomplete outcome data||Selective outcome reporting||Other|
|Bouzo||Unclear||unclear||Low risk||Low risk||Low risk||unclear||High Risk|
|Farup||Unclear||unclear||Low risk||Low risk||Low risk||unclear||Low risk|
|Khuroo||Unclear||unclear||Low risk||High risk||Low risk||Low risk||Low risk|
|Cohort and case–control studies||Representativeness (1)||Selection of non-exposed (1)||Ascertainment of exposure (1)||Outcome of interest not present at start of study (1)||Comparability (2)||Assessment of outcome (1)||Follow up long enough and adequate (2)|
|Druart-Blazy||1||1||1||1||0 + 0||1||1|
|Cicala||1||1||1||1||0 + 0||0||1|
|Bennett||1||1||1||1||1 + 0||1||1|
The Cochrane risk of bias score was used for the 3 RCTs included and the Newcastle–Ottawa score for the cohort and case control studies. The score allocated for each component of the relevant validated scoring systems can be seen demonstrating which part of the study methodology was open to potential bias.
The cohort studies by Druart-Blazy et al.19 and Cicala et al.20 scored 6 and 5, respectively on the Newcastle–Ottawa risk of bias scale. Neither controlled for group differences at baseline and all failed to score for comparability. Notably nine patients in the non-opiate group in the study by Druart-Blazy et al.19 had an endoscopic sphincterotomy versus none in opiate group. As the cohorts were not matched at baseline, it is likely that the opiate group had more severe symptoms, necessitating their opiate use. In the study by Cicala et al.20patients were allocated to conservative treatment over sphincterotomy if they refused the intervention. This would create a self-selecting group with significant comparability bias.
The case–control study by Bennett et al.21 looked for association between psychosocial variables and SOD, and scored 7 on the Newcastle–Ottawa risk of bias scale. The study group’s baseline characteristics (age, representation across SOD classifications) were all similar suggesting appropriate controls were made for comparability.
This systematic review collates the findings of nearly three decades of research on PCS. A key finding of this review is the spectrum of aetiologies that contribute to PCS. In the first three years post cholecystectomy the most frequent diagnoses are peptic ulcer disease, GORD and hiatal hernias. Beyond three years, the most common diagnoses relate to ductal stones and microlithiasis although nearly a third of cases will not have a conclusive diagnosis. The frequency of SOD is low, accounting for fewer than 50% PCS symptoms even in populations that have had other causes of PCS excluded. The literature exploring management of post cholecystectomy syndrome is scarce and heterogeneous, and the studies that do exist are limited by small numbers of participants or flawed methodology.
The definition of post cholecystectomy syndrome is broad and variable. The term encompasses a selection of non-specific symptoms, which may or may not be related to or exacerbated by the surgical removal of the gallbladder. Definitions vary regarding the timing of symptoms after cholecystectomy and consensus is lacking. Timing of symptoms is important as arguably the longer the delay between surgery and the development of further symptoms, the less likely it is that these can be attributed to cholecystectomy. We have demonstrated that within three years of surgery, PCS symptoms are mostly caused by non-biliary pathologies. Beyond three years, biliary-related pathologies do account for most symptoms, but these are not related to cholecystectomy. Therefore if a patient presents with biliary symptoms post-cholecystectomy the term PCS should be used to re-start the diagnostic process as opposed to being a diagnosis it its own right. Ultimately the term PCS may become redundant.
In the studies collated in this review, it was unclear whether the pain was new in onset, followed surgery, or was persistent from the pre-operative period. The latter could suggest mis-diagnosis of the original symptoms. This dilemma was highlighted by Rogy et al.,29 who suggested that the most common cause for PCS was overlooked extra-biliary disorders.
Describing the most prevalent causes of PCS is useful to guide the diagnostic process for patients presenting with symptoms post cholecystectomy, and ultimately hasten the identification and treatment of the cause. Notably the most common causes are easily identifiable on endoscopy or routine imaging. We found that beyond three years post cholecystectomy the types of aetiology tended to change. Aetiologies moved from gastric causes towards an increased incidence of biliary lithiasis or pathologies that were in themselves risk factors for stone formation, such as biliary strictures and stenosis. This temporal pattern in aetiology has also been demonstrated by Skalicky,50 who reported that PCS patients may present with ‘early atypical symptoms’ of PCS or late symptoms as much as 3–5 years following surgery, with the greater proportion of reported PCS occurring in the latter group.38 Other studies describe retained biliary stones, sphincter of Oddi dysfunction and biliary stenosis as the most frequent causes of PCS24, 39, 40 although we found that these aetiologies were most prevalent several years after cholecystectomy. These findings are explicable by the changes in biliary physiology following removal of the gallbladder. The loss of the reservoir function of the gallbladder is partly compensated for by dilation of the biliary tree, which predisposes to biliary stasis, the formation of stones and possible infection.41Post cholecystectomy patients have also been shown to have an increased CBD pressure, which may partly contribute to the development of PCS.42
In this systematic review most studies used a range of blood results, radiologyand endoscopy to investigate and diagnose the cause of PCS. It is beyond the scope of this paper to synthesize the efficacy and sensitivity of diagnostic methods but several studies did attempt to assess this. Filip et al.25 reported elevated serum bilirubin and alkaline phosphatase (ALP) in combination with abdominal ultrasound accurately predicted PCS in 90% cases. In the present review we have found that up to 50% patients had PCS symptoms despite normal investigations though.
Several studies have tried to identify predictive factors for PCS. Bisgaard et al.6found that the incidence of PCS was higher in patients who had experienced intense acute post-operative pain. This has been shown in other studies of chronic pain following surgery.43 However, patient’s risk of experiencing acute post-operative pain could not be predicted by their pain threshold pre-operatively, as assessed by a nociceptive stimulus i.e. cold pressor test.6
Stefaniak et al.51 proposed that PCS had a psychosocial influence and those without a social network for support and increased remuneration were more likely to somatise symptoms. Bennett went on to describe groupings of psychosocial and gender factors which together differentiated PCS patients with sphincter of Oddi dyskinesia (but not stenosis) from PCS patients with normal sphincter of Oddi motor activity, suggesting an association with functional PCS and social status.
Studies examining the management of PCS are few in number. We identified three studies of pharmacological intervention, all examining cisapride or nifedipine. In 2000 the Medicines Control Agency withdrew the licence for cisapride in the UK following a review by the Committee on Safety of Medicines that reported rare but serious side effects, including long QT syndrome and other cardiac arrhythmias, some of which proved fatal.44Calcium channel antagonists inhibit sphincter of Oddi contraction.45 Studies of sublingual nifedipine at endoscopic manometry demonstrate a reduction in basal Sphincter of Oddi pressure and fewer phasic contractions.46 The RCT by Khuroo et al.18 demonstrated lower pain scores and fewer unscheduled visits to secondary care with regular nifedipine. This was a small trial (28 participants), and the use of nifedipine in PCS warrants further research. Additionally whether nifedipine has a different risk and side effect profile in a population with normal blood pressures requires further assessment.
Sphincter of Oddi pressures can also be attenuated using endoscopic sphincterotomy. Endoscopic sphincterotomy is thought to be the treatment of choice for the management of SOD when basal pressures are elevated (>40 mmHg),47 and can assist with retrieval of bile duct stones. In the current review we found no randomised evidence supporting its efficacy. Evidence was confined to a cohort study of 22 participants, finding that endoscopic sphincterotomy was associated with a reduction in recurrent PCS symptoms. This study did not report adverse events however, which is important given risks of this procedure: pancreatitis (5%), haemorrhage (2%) and cholangitis(1%).48 The other study identified was a case series, which reported improvement in symptoms,27, 32 and procedure related complication rates of 5.7%,33 consistent with the literature.
There are several limitations to this systematic review. The broad definition of PCS makes the identification of eligible studies difficult. Furthermore the heterogeneity of study populations precludes meta-analysis, significantly hindering the power and generalizability of any conclusions. The quality of studies was mostly poor and underpowered. Modes of investigation also varied greatly between the studies, making synthesis of aetiological data less reliable. To mitigate this we re-classified many of the authors’ classifications into broad groups. The culmination of these limitations makes the derivation of any definitive conclusions that can aid clinical decision making extremely difficult. The search strategy was limited to full English papers and we would recommend that any future revisions of this systematic review should search and include relevant papers from publications of all languages.
This systematic review acts as a caution. Highlighting the lack of a universally agreed clinical definition for post cholecystectomy syndrome and the paucity of robust studies to help define or better understand this poorly understood syndrome. If this syndrome is to remain in medical vocabulary there is an imminent need for studies with a robust methodology and adequately powered to clarify the remit of this syndrome and ensure consistency in its use and applicability.
Post cholecystectomy syndrome is used to describe a collection of symptoms experienced by patients following cholecystectomy, many of which can be attributed to causes beyond the biliary system. The symptoms experienced are not collectively representative of a single disease or abnormality so referencing them as a syndrome may be inaccurate. Whether some of the symptoms are truly a result of gallbladder removal is questionable. In the immediate period following laparoscopic cholecystectomy many patients presenting with pain will have GORD or peptic ulcer disease. Physicians should exclude these causes before pursuing specialised investigations of the biliary system. In the later period, ductal stones become more common, but a third of patients will have no cause identified. Treatment options for this latter group of patients are limited.
Miss Isherwood and Mr Khanna designed the hypothesis.
All authors were involved in review of abstracts and full papers, as an independent reviewer or arbitrator.
Miss Isherwood and Miss Oakland extracted and interpreted the data.
Miss Isherwood drafted the manuscript and Miss Oakland and Mr Khanna provided critical revisions.
The authors have no conflicts of interest or financial ties to disclose.
Ethical approval statement
For this type of study formal consent is not required.
Informed consent statement
Does not apply.
The Authors have no acknowledgements to make.
Appendix A. Supplementary data
The following are the supplementary data related to this article: