Readmissions after cholecystectomy in a tertiary UK centre: Incidence, causes and burden.

CONTEXT: Although cholecystectomy is a widely performed procedure, post-operative readmissions place a heavy burden on healthcare facilities. AIMS: This study assesses the incidence, causes and burden of 30-day readmissions after cholecystectomy in a tertiary UK centre. SETTINGS AND
DESIGN: This study was conducted at a university hospital, and the study design involves retrospective cohort study.
MATERIALS AND METHODS: Information was obtained from our prospectively maintained database and hospital's computerised records. STATISTICAL ANALYSIS: The encounters are expressed in numbers and percentages. The hospital stay, body mass index and age are expressed in mean, standard deviation (SD), min-max and median. Microsoft Excel® was used to calculate the means, SD, min-max and median.
RESULTS: Out of the 1140 cholecystectomies performed over this time, there were 75 true readmissions and 29 revisits; thus, the actual readmission rate is 6.58%. Non-specific abdominal pain ± deranged liver function test (LFT) is the most common cause of readmissions/ revisits in (38; 36.54%) cases, followed by (18; 17.31%) wound infections and (12; 11.54%) collections/bile leaks/abscess. This cost the centre 93 scans, 30 procedures and 295 days of hospital stay.
CONCLUSIONS: Non-specific abdominal pain ± deranged LFT is the most common cause of readmissions/revisits in the centre. Readmissions after a cholecystectomy are a significant encumbrance.

INTRODUCTION

A cholecystectomy, which is the gold standard for the management of gallstone disease, is one of the most widely performed surgical procedures, with over 1,000,000 procedures performed annually in the United States (US) and over 50,000 in the United Kingdom (UK).[1234] Despite it being an established operation, post-operative complications necessitating readmission still occur, and this places a heavy burden on healthcare providers and impacts the patient's satisfaction with the service provided, as well as causing morbidity and mortality.[567]

Hospital readmission has become a key performance indicator and care quality measure.[58] In the UK, the latest Association of Upper Gastrointestinal Surgery of Great Britain and Ireland (AUGIS) commission guidelines[9] set the benchmark for readmission within 30 days after a cholecystectomy as <10%. Readmissions after a cholecystectomy for non-surgical or anaesthesia-related reasons such as post-operative nausea, vomiting and non-specific abdominal pain may cause this to increase to >40%.[810]

The aim of this study was to assess the incidence and causes of readmissions after a cholecystectomy in a tertiary hospital in the UK that specialises in benign upper gastrointestinal (UGI) surgery. Moreover, light is shed on the burden of these readmissions in terms of the investigations and interventions required and the overall hospital stay necessitated.

MATERIALS AND METHODS

Study design

A retrospective cohort study was conducted using the centre's electronic database, which contains data on all 30-day readmissions after cholecystectomy procedures.

Setting

The setting of this study was a large benign UGI surgery unit in a university-affiliated tertiary care hospital in the UK.

Participants

This study's participant population comprised all patients who were readmitted after a cholecystectomy over a 2-year period, specifically from 1 April 2018 to 31 March 2020.

Variables examined

The demographic and pre-operative data that were collected comprised age, gender, body mass index (BMI) and diabetes mellitus status. The operative data that were collected comprised mode of surgery (laparoscopic or open), the admission rate from day case surgery and the surgeon's level of experience (consultant or trainee).

The readmissions episodes were thoroughly studied to assess the cause of readmission, investigations obtained and their findings, interventions required and the hospital stay on readmission.

Data source/bias

The data were collected from the hospital's computerised records and patient's case notes as and when necessary.

Study size

The total number of cholecystectomies performed over the time interval was 1140. Of these, 104 patients came back to the hospital after their cholecystectomy, with 75 of these being true readmissions and 29 being revisits.

Definitions

True readmission

It is defined as returning to the hospital within 30 days of discharge after a cholecystectomy and staying overnight or longer.

Revisit

It is defined as returning to the hospital within 30 days of discharge after a cholecystectomy without staying overnight (<24 h stay).

Emergency cholecystectomy

Cholecystectomy performed at the index admission or within 2 weeks of presentation with an acute condition.

Elective cholecystectomy

Cholecystectomy performed due to biliary colic or more than 2 weeks after presentation with an acute condition.

Statistical analysis

The encounters are expressed in numbers and percentages. The hospital stay, BMI and age are expressed in mean, standard deviation (SD), min-max and median. Microsoft Excel® was used to calculate the means, SD, min-max and median.

RESULTS

Over the 2 years, from 1 April 2018 to 31 March 2020, the total number of cholecystectomies performed was 1140, with 1122 being laparoscopic (98.42%), nine nine being open (0.79%) and nine being laparoscopic converted to open (0.79%).

Emergency cholecystectomies were performed on 287 patients (25.17%), of which 280 were laparoscopic (97.56%), four were open (1.39%) and three were laparoscopic converted to open (1.05%), with 13 being day cases (4.53%). In addition, 853 procedures were performed on an elective basis (74.82%), of which 842 were laparoscopic (98.71%), five were open (0.59%) and six were laparoscopic converted to open (0.7%), with 635 being day cases (74.44%). The total number of day- case procedures was 648 (56.84%).

In terms of pre-operative endoscopic retrograde cholangiopancreatography (ERCP), 12 ERCPs were done. Nine of these were an emergency and three of these were on an elective basis. An on-table cholangiogram was performed 111 times, with 82 of these being on elective cases and 29 taking place in an emergency setting.

In total, 104 readmissions were recorded on the database, yielding a readmission rate of 9.12%. Of these, 71 were female and 33 were male, the mean age was 49, the SD was ± 20, the min-max was 17–87 and the median was 47. An analysis of these 104 patients revealed that 75 of these were true readmissions and 29 were revisits, i.e., an assessment in ambulatory care without staying overnight, yielding an actual readmission rate of 75/1140 (6.58%). An analysis of the entire group of revisits and true readmissions revealed that 100 of the procedures were performed laparoscopically (96.15%), three were laparoscopic converted to open (2.88%) and one was an open cholecystectomy (0.96%).

A review of the cholecystectomies performed on these patients revealed that 51 of the procedures were a day case surgery where the patient was discharged from the day case (49%), 15 involved the patient being admitted from a day case (14.42%) and 38 were done on inpatients after an emergency admission (36.54%). Consultants performed 55 of the procedures (52.88%) and 49 were done by trainees (47.12%).

Table 1 shows the total causes of true readmissions/revisits. Non-specific abdominal pain ± deranged liver function test (LFT) was the most common cause in 38 cases (36.54%), which was followed by wound infections in 18 cases (17.31%) and collections/bile leaks/abscess in 12 cases (11.54%). There were six missed stones with collections and six planned reviews, constituting 5.77% each.

Table 1

Reasons for true readmissions and revisits (n=104)

Cause	n (%)
Non-specific abdominal pain±deranged LFT	38 (36.54)
Wound infection	18 (17.31)*
Collection/bile leaks/abscess	12 (11.54)*
Missed stones + collections	6 (5.77)*
Planned	6 (5.77)
Constipation	4 (3.85)
Urine retention	2 (1.92)
Collection+wound infection	2 (1.92)*
Vomiting	2 (1.92)
Liver and port-site haematoma	1 (0.96)*
Pancreatitis	1 (0.96)
Superficial vein thrombosis	1 (0.96)
Bowel obstruction	1 (0.96)#
Nausea	1 (0.96)
Pyelonephritis	1 (0.96)
Leaking drain	1 (0.96)*
Port site bruising and pain	1 (0.96)*
UTI	1 (0.96)
Port site hernia+wound infection	1 (0.96)*
Infected pseudocyst	1 (0.96)
Night sweat	1 (0.96)
CHD stricture+bile leak	1 (0.96)*
Wound bleeding	1 (0.96)*

*Surgical complications, #Not related to cholecystectomy. LFT: Liver function tests, UTI: Urinary tract infections, CHD: Common hepatic duct

The results were broken down further to investigate the reasons for the true readmissions and revisits separately [Table 2]. In the revisit group (n = 29), the most common reason for revisits was wound infection in 12 cases (41.38%), which was followed by non-specific abdominal pain ± deranged LFT in nine cases (31.03%) and planned reviews in four cases (13.79%). However, in the true readmission group (n = 75), the most common reason for a readmission was non specific abdominal pain ± deranged LFT in 29 cases (38.67%), which was followed by collection/bile leaks/abscess in 12 cases(16%), wound infection in six cases (8%) and missed stones with collections in six cases (8%).

Table 2

Breakdown of reasons for revisits and true readmissions

Reason	n (%)
Revisits
Wound infection	12 (41.38)
Non-specific abdominal pain±deranged LFT	9 (31.03)
Planned (2 OGD – 2 review of bloods)	4 (13.79)
Constipation	1 (3.45)
Port site bruising and pain	1 (3.45)
Night sweat	1 (3.45)
UTI	1 (3.45)
Total	29
True readmissions
Non-specific abdominal pain±deranged LFT	29 (38.67)
Collection/bile leaks/abscess	12 (16.00)
Wound infection	6 (8.00)
Missed stones + collections	6 (8.00)
Constipation	3 (4.00)
Planned	2 (2.67)
Urine retention	2 (2.67)
Collection+wound infection	2 (2.67)
Vomiting	2 (2.67)
Liver and port-site haematoma	1 (1.33)
Pancreatitis	1 (1.33)
Superficial vein thrombosis	1 (1.33)
Bowel obstruction	1 (1.33)
Nausea	1 (1.33)
Pyelonephritis	1 (1.33)
Leaking drain	1 (1.33)
Port site hernia+wound infection	1 (1.33)
Infected pseudocyst	1 (1.33)
CHD stricture+bile leak	1 (1.33)
Wound bleeding	1 (1.33)
Total	75

LFT: Liver function tests, UTI: Urinary tract infections, CHD: Common hepatic duct, OGD: Oesophago-gastro-duodenoscopy

The secondary outcome of this study was to estimate the burden of these readmissions and revisits in terms of the required investigations, interventions and hospital stays [Table 3]. Ninety-three imaging studies were performed, comprising 48 computerised tomography (CT) scans, 24 ultrasounds (USs) and 19 magnetic resonance cholangiopancreatographies (MRCPs). Moreover, nine ERCPs, eight laparoscopic drainage and washouts, four wound incisions drainage (IandD) and debridement were required, in addition to three US-guided drainage of collections.

Table 3

Burden: imaging and interventions

	n
Imaging modality (n=93)
CT	48
US	24
MRCP	19
US Doppler	1
MRI	1
Interventions (n=30)
ERCP	9
Lap drainage+washout+drains	8
Wound I and D+debridement	4
US drainage	3
Lap CBD exploration	1
Right hemicolectomy	1
Lap hernia repair	1
Lap excision of cystic duct stump	1
Referred to HBP	1
Dilatation of drain tract and drainage of bile	1

CT: Computerised tomography, US: Ultrasound, MRCP: Magnetic resonance cholangiopancreatography, I and D: Incisions and drainage, MRI: Magnetic resonance imaging, ERCP: Endoscopic retrograde cholangiopancreatography, CBD: Common bile duct, HBP: Hepatobiliary and pancreas surgery

The total number of hospital stay days for the true readmission group (n = 75) was 295 days with a mean of four, SD of ±4, min-max of 1–17 and a median of 2 days.

A further analysis focusing on the most common causes showed that there were 38 cases of non-specific abdominal pain ± deranged LFT, of which 29 were true readmissions and nine were revisits. Four USs had to be done for the nine revisits. However, for the 29 patients who were readmitted, 13 CTs, 12 USs, 10 MRCPs and one magnetic resonance imaging scan were required. The mean hospital stay for these patients was 2 days with an SD of ± 1.61, a min-max of 1–7 and a median of 2 days. The total number of hospital admission days was 65 days. No surgical interventions were required except for one case involving the laparoscopic excision of the cystic duct stump.

The second most common reason for readmissions/ revisits was wound infection in 18 cases, of which 12 were revisits that were managed in ambulatory care and six were true readmissions. The interventions required for this group included two IandDs and one wound debridement for the readmitted group (3/6), in addition to one IandD for the revisit group (1/12). The mean hospital stay for the admitted group with wound infections was 2.16 with an SD of ±1.83, a min-max of 1–5 and a median of 1 day and a total of 13 days of hospital stay.

Regarding the risk factors for wound infection, only one out of the 18 patients was diabetic compared to nine out of 86 for the other complications. The mean BMI for the wound infection group was 32, with it having an SD ±6.17, a min-max of 23–48 and a median of 33 compared to a mean BMI of 30, an SD of ±7, a min-max of 18–65 and a median of 29 for the other complications.

DISCUSSION

Although the centre's readmission rate (6.58%) is in line with the local benchmark,[9] these results demonstrate the heavy burden of readmissions after gallstone surgery. A recent meta-analysis[11] performed on 44 studies involving 1,573,715 laparoscopic cholecystectomies from 25 countries found a 30-day readmission rate of 3.3%, ranging from 0% to 11.7%. However, this meta-analysis found a higher readmission rate of 7.7% in 20 European studies compared to 3.6% in eight North American studies.

This meta-analysis found that surgical complications accounted for 76% of the reported readmissions, which were mainly due to bile duct complications (33%) followed by wound infection (17%) and nausea and vomiting (9%). Pain and cardiorespiratory complications accounted for 15% and 8%, respectively. The results of the current study show that non-specific abdominal pain was the most common reason for true readmissions, accounting for 38.67%, followed by the other surgical complications, namely collection/bile leaks/abscess (16%), wound infection (8%) and missed stones + collections (8%).

Other UK studies[412] report similar readmission rates to the current study, for instance 6.6% and 6.7%, with non-specific abdominal pain being the leading cause for readmission, ranging from 36% to 45.5% of readmissions. In a multicentre study in the US,[13] out of 230,745 laparoscopic cholecystectomies performed in 890 facilities in three different states, the rate of 30-day readmission was found to be 2.02%, and the most common diagnoses on readmission were surgical complications, post-operative pain, infection and nausea or vomiting. Another recent US study[14] found a readmission rate of 2.8%, with only 14.07% of the readmissions due to abdominal pain.

This could, in part, explain the higher readmission rates reported in the UK compared to North America, as most of the readmissions in the UK are due to non-specific symptoms such as pain, nausea and vomiting without significant findings. This may reflect a difference in practice between the UK and the US, with definitive diagnostic imaging studies being performed in the UK after the decision to admit rendering the readmission rate falsely higher. In other words, a normal CT scan in an emergency department or at surgical ambulatory facilities would have avoided unnecessary readmissions and reduced the burden on the National Health Service (NHS).

The analysis of the current study's findings shows that surgical complications collectively accounted for 44 (42.31%) of the total revisits and true readmissions [as indicated in Table 1]. Despite this, non-surgical causes of readmissions and revisits were still more prevalent, including post-operative nausea, vomiting, constipation and other presentations of non-specific pain.

There was only one case of common hepatic duct stricture with bile leak. Although it is only once case out of 1140 cholecystectomies performed at the centre, the centre's benchmark[9] is zero bile duct injuries.

The secondary outcome of this study was to assess the burden that readmissions place on NHS facilities, which has been proven to be heavy. In a single centre over only 2 years, 104 readmissions/revisits necessitated 93 imaging studies, 30 procedures [Table 3] and 295 days of hospital stay.

As a possible way to reduce this burden, the causes of readmission that may be avoidable were looked at. Out of the 38 cases of non-specific pain, 29 were readmitted. The total number of hospital stay days for these 29 cases was 65 days, and 36 imaging studies were done in addition to the four USs for the nine patients who were attended to in the ambulatory surgical unit without being admitted. Here, an initial proper assessment at the surgical ambulatory care unit and emergency department with the use of appropriate scans as needed to rule out real surgical complications is recommended. However, great caution should be exercised so as not to discharge patients presenting with sepsis and clear indications for readmission. On most occasions, the management of wound infections was on an outpatient basis (12/18 cases). Finally, we found that 11 cases presented with constipation, nausea, vomiting, night sweats, urinary tract infections and urine retention. Eight of these were readmitted and three were managed in the ambulatory unit.

Most of these patients could probably have been managed properly in the ambulatory unit without the need for readmission. Proper antiemetic medications with effective multimodal analgesia in the post-operative period and after discharge would help reduce these unnecessary non-surgical readmissions after gallbladder surgery.[15]

Study limitations

The retrospective nature of the study design and its focus on the readmitted group did not enable the investigation of the risk factors and predictors of readmission.

Only the 30-day readmission rate was assessed. The 90-day readmission rate could have been studied to obtain a clearer picture of readmission due to delayed or long-term complications.

None of the operative scoring systems[161718] that could have helped predict the outcome and would have aided in the comparison of the patients’ operative severity and grade are routinely used.

Generalisability

Readmission rates that are similar to other studies conducted in the UK and higher than the rates in North America were observed, with the most common reason for readmission being non-specific abdominal pain alongside other non-surgical complaints, such as nausea, vomiting and constipation. It is worth mentioning that most of the studies from North America report lower rates of non-specific pain as a cause of readmission after gallbladder surgery. It is clear to the authors that unnecessary readmissions after cholecystectomy in the UK represent a challenge, and these non-surgical complaints should be managed on an outpatient basis given the fact that most of the UK's hospitals have surgical ambulatory or same day care units.

CONCLUSIONS

The readmission rate after a cholecystectomy in the tertiary centre is 6.58%, which is in line with the local benchmark. Non-specific abdominal pain ± deranged LFT is the most common cause of true readmissions in the centre at 38.67%. Readmission after a cholecystectomy places a huge burden on healthcare facilities. Post-operative complaints due to non-surgical causes such as non-specific abdominal pain, vomiting, nausea and constipation can be investigated and managed on an outpatient basis and in ambulatory care units to reduce the burden of these readmissions.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.