Introduction
Surgical treatment of symptomatic gallstones through open or laparoscopic cholecystectomy is one of the most frequent procedures performed throughout the world1,2. Although most of cholecystectomies are performed laparoscopically and even in an outpatient setting, hospital readmission can occur in a substantial number of the cases3. A quality standard of < 10% readmission rate after cholecystectomy within the post-operative 30 days has been also recommended4. After any medical or surgical treatments, hospital readmission has been regarded as a measure of hospital and surgical quality of care1,5.
In the previous studies, avoidance from readmissions has been questioned due to different patients' conditions and factors3. Besides, it has been speculated that hospital readmissions may result from complications that were not recognized before discharge or inadequate discharge planning6. Thus, efforts have been developed to reduce the rate of such admissions and to improve delivery of care for patients undergoing cholecystectomy considering the medical importance and cost of unplanned readmissions following cholecystectomy6,7.
The incidence, risk factors, and causes of unplanned hospital readmission after cholecystectomy have been analyzed in the previous studies1,3-5,8-10. The rate has varied from 2.4% to 6.7%, with an average of 3.3% depending on the different settings of cholecystectomy7. Identifying the factors responsible for unplanned readmissions may help physicians develop pre- and post-operative strategies to avoid such admissions.
Although most cholecystectomies have been performed laparoscopically, there are a limited number of studies, in which open cholecystectomy is considered for the calculation of readmission rates following cholecystectomy1,5. Thus, analysis of all laparoscopic, open, and converted cholecystectomies regarding hospital readmissions may be used to develop a discharge strategy to define risky patients for hospital readmissions.
In the present study, we aimed to investigate the rate of unplanned readmissions after cholecystectomy and evaluate the reasons and outcomes.
Materials and methods
Study
This study was a retrospective analysis of all consecutive patients who underwent open and laparoscopic cholecystectomy for benign biliary pathology between June 2019 and March 2020 at Bagcilar Training and Research Hospital, Istanbul, Turkey regarding hospital readmissions within the post-operative first 90 days after the procedure. The Local Ethical Committee (Bagcilar Training and Research Hospital, May 15, 2020, Approval number: 2020.05.1.02.034) approved the study. The researchers were in agreement with the Declaration Helsinki for the study. Written consent could not be taken from the patients due to the retrospective design of the study and deidentified nature of data used for the analyses.
Patients
All patients with a pre-operative diagnosis of “cholelithiasis,” “cholesterolosis,” “chronic cholecycstitis,” and “acute cholecystitis” were searched using the hospital information system. Patients who were younger than 18 years, patients with coexisting common bile duct pathologies, current oncological diseases, and cholecystectomy combined with other surgical operations were not included in the study. All laparoscopic procedures were performed by the classical 3-or 4-port technique4. As an institutional policy, ambulatory, single incision laparoscopic, or robotic surgery were not used for cholecystectomy.
Variables including patients' demographics (age, sex, body mass index (BMI), comorbidities, the American Society of Anesthesiologists (ASA) score, previous biliary pathologies necessitating interventions (endoscopic retrograde cholangiopancreatography (ERCP), and other percutaneous interventions), imaging findings, operative findings (type of presentation as elective or emergent, type of operation as open, laparoscopic and conversion, operation time, and intraoperative complications), and post-operative features (length of hospital stay and post-operative complications) were recorded. The operation time and length of hospital stay of all patients were grouped using the median value of each variable as cut off. The final gallbladder pathology was also recorded.
Besides, the reasons for hospital readmissions were classified, according to Rana et al.9:
The length of hospital stay during readmission, the treatment modalities, and outcomes was evaluated.
Outcomes
The rate and reasons for hospital readmissions within the post-operative first 90 days after cholecystectomy were the primary outcomes of the study. The risk factors of hospital readmissions were the secondary outcomes. Hospital readmission was defined as any hospitalization with a duration of at least 24 h at the emergency service or general surgery wards. If a patient had more than one hospital readmissions with the same complaints, repeated admissions were ignored. However, in a case of more than one hospital readmissions with different complaints, each admission was regarded as a new admission. Hospital admissions that were unrelated to the index operation were not included in the study.
Statistical analysis
Statistical analysis was performed using a statistical package (SPSS 15.0, Chicago, IL, USA). The normality of distribution of continuous variables was assessed using the Kolmogorov– Smirnov test. Continuous variables with and without normal distribution were presented as mean ± standard deviation and median with interquartile range (IQR) of 25-75%. Categorical variables were presented with frequencies with percentages. The Pearson Chi-square and Fisher's exact tests were used for categorical variables. The t-test and one-way analysis of variance (ANOVA) were used to compare the continuous variables with the normal distribution. The Mann–Whitney U-test was applied to compare continuous variables without normal distribution. p < 0.05 was considered statistically significant.
Results
There were 601 patients with a mean age of 53.2 ± 12.4 years. The baseline characteristics of the patients are summarized in table 1. Female-to-male ratio was 2. Most of the patients were female (66.6%), had BMI between 25.0 and 34.9 kg/m2 (70.7%), had ASA scores of 1 and 2 (93.4%), and had no coexisting disease (59.7%). Diabetes mellitus was the most common coexisting disease seen in 76 patients (12.6%).
Variable | Overall (n = 601) | Group 1 (n = 37) | Group 2 (n = 564) | p |
---|---|---|---|---|
Age (year)† | 53.2 ± 12.4 | 51.6 ± 13.4 | 53.3 ± 12.3 | 0.684 |
Sex‡ | ||||
Female | 400 (66.6) | 26 (70.3) | 374 (66.3) | 0.721 |
Male | 201 (33.4) | 11 (29.7) | 190 (33.7) | |
BMI (kg/m2)† | 30.9 ± 5.0 | 35.1 ± 4.6 | 30.6 ± 4.9 | 0.001 |
BMI groups‡ | ||||
Underweight (< 18.5) | 2 (0.3) | 0 (0) | 2 (0.4) | < 0.001 |
Normal (18.5-24.9) | 63 (10.5) | 0 (0) | 63 (11.2) | |
Overweight (25-29.9) | 202 (33.6) | 6 (16.2) | 196 (34.8) | |
Grade I obesity (30-34.9) | 223 (37.1) | 10 (27.0) | 213 (37.8) | |
Grade II obesity (35-39.9) | 89 (14.8) | 17 (45.9) | 72 (12.8) | |
Grade III obesity (≥40) | 22 (3.7) | 4 (10.8) | 18 (3.2) | |
Smoking‡ | 39 (6.5) | 2 (5.4) | 37 (6.6) | 1.0 |
Number of coexisting disease‡ | ||||
None | 359 (59.7) | 23 (62.2) | 336 (59.6) | |
1 | 162 (27.0) | 6 (16.2) | 156 (27.7) | |
2 | 73 (12.1) | 6 (16.2) | 67 (11.9) | |
≥3 | 7 (1.2) | 2 (5.4) | 5 (0.9) | 0.039 |
ASA grade‡ | ||||
1 | 301 (50.1) | 17 (45.9) | 284 (50.4) | 0.008 |
2 | 620 (43.3) | 13 (35.1) | 247 (43.8) | |
3 | 40 (6.7) | 7 (18.9) | 33 (5.9) | |
Diabetes mellitus‡ | 76 (12.6) | 6 (16.2) | 70 (12.4) | 0.449 |
CAD-CHF‡ | 41 (6.8) | 4 (10.8) | 37 (6.6) | 0.307 |
COPD‡ | 38 (6.3) | 5 (13.5) | 33 (5.9) | 0.075 |
†mean ± standard deviation,
‡n (%)
BMI: body mass index, ASA: the American Society of Anesthesiologists, CAD-CHF: coronary artery disease-congestive heart failure, COPD: chronic obstructive pulmonary disease.
Pre-operative US revealed that cholelithiasis was detected in 541 patients (90%). Increased wall thickness was seen in 75 patients (12.5%). Pre-operative interventions as ERCP and PC were performed in 23 patients (3.8%) (Table 2).
Variable | Overall (n = 601) | Group 1 (n = 37) | Group 2 (n = 564) | p |
---|---|---|---|---|
Cholelithiasis‡ | 541 (90) | 32 (86.5) | 509 (90.2) | 0.402 |
GB polyps‡ | 20 (3.3) | 2 (5.4) | 18 (3.2) | 0.352 |
Increased wall thickness‡ | 75 (12.5) | 4 (10.8) | 61 (10.8) | 1.0 |
Scleroatrophic GB‡ | 11 (1.8) | 2 (5.4) | 9 (1.6) | 0.143 |
Pre-operative interventions‡ | 23 (3.8) | 1 (2.7) | 22 (3.9) | 1.0 |
ERCP | 21 (3.5) | 1 (2.7) | 20 (3.5) | NA |
PC | 4 (0.7) | 0 (0) | 4 (0.7) | NA |
‡n (%)
GB: the gallbladder, ERCP: endoscopic retrograde cholangiopancreatography,
PC: percutaneous cholecystectomy.
Operative features are detailed in table 3. In 563 patients (93.7), the operation was performed as elective cholecystectomy. Laparoscopic cholecystectomy was the most common type of surgery in the study group (89.2%). The median length of operation time was 55 min, and in 310 patients (51.6%), the length of the operation was longer than 55 min. There were 47 intraoperative complications (7.8%) as bleeding and biliary injury seen in 42 and five patients, respectively. The bleedings from the cystic artery, liver bed, and the entry points for trocars were seen in five, 31, and six patients, respectively. The patients with intraoperative biliary injury were converted to open surgery. Primary repair over the T tube and hepaticojejunostomy was performed in two and three patients, respectively.
Variable | Overall (n = 601) | Group 1 (n = 37) | Group 2 (n = 564) | p |
---|---|---|---|---|
Presentation‡ | ||||
Elective | 563 (93.7) | 35 (94.6) | 528 (93.6) | 1.0 |
Emergent | 38 (6.3) | 2 (5.4) | 36 (6.4) | |
Type of surgery‡ | ||||
Laparoscopic | 536 (89.2) | 29 (78.4) | 507 (89.9) | 0.002 |
Open | 21 (3.5) | 0 (0) | 21 (3.7) | |
Converted | 44 (7.3) | 8 (21.6) | 36 (6.4) | |
Operation time (min)β | 55 (40 60) | 55 (35 60) | 55 (40 60) | 0.875 |
Operation time group‡ | ||||
≤ 55 min | 291 (48.4) | 18 (48.6) | 291 (51.6) | 1.0 |
> 55 min | 310 (51.6) | 19 (51.4) | 273 (48.4) | |
Intraoperative complications‡ | ||||
None | 556 (92.5) | 28 (75.7) | 528 (93.6) | < 0.001 |
Bleeding | 42 (6.9) | 8 (21.6) | 34 (6.0) | |
Biliary injury | 5 (0.78) | 3 (7.1) | 2 (0.4) | |
Use of drain‡ | ||||
Yes | 111 (18.5) | 15 (40.5) | 96 (17.0) | 0.001 |
No | 490 (81.5) | 22 (59.5) | 468 (83.0) | |
LOS (day)β | 1 (1-1) | 1 (1-2) | 1 (1-1) | 0.013 |
LOS group‡ | ||||
≤ 1 | 468 (77.9) | 23 (62.2) | 445 (78.9) | 0.024 |
> 1 | 133 (22.1) | 14 (37.8) | 119 (21.1) |
*n (%),
βmedian (interquartile range)
LOS: length of hospital stay.
A total of 37 patients (6.16%) readmitted during the post-operative first 90 days. There were significant differences between the BMI of the patients in Group 1 and Group 2. The BMI of the patients in Group 1 was significantly higher than that of Group 2 (p = 0.001). Patients with Grade II and Grade III obesity were more frequently seen in Group 1 (p < 0.001). As the number of coexisting diseases increased, the rate of readmission significantly increased (p = 0.039). There were more patients with ASA Grade III in Group 1 compared within Group 2 (18.9% vs. 5.9%) (p = 0.008). There were no significant differences in age, sex, smoking history, type of coexisting diseases, and imaging features between the groups (p > 0.05 for all) (Tables 1 and 2). The patients with conversion and intraoperative complications were more frequently readmitted (p = 0.002 and p < 0.001) (Table 3). The use of drain was performed in 17.0% of the patients in Group 2, whereas 40.5% of the patients in Group 1 received an intraoperative drain during their index operation (p = 0.001).
Type of presentation and operation time revealed no significant difference between the groups (p > 0.05 for both). Although the median length of hospital stay was 1 day in both groups, there was a significant difference in the length of hospital stay (p = 0.013). The patients with a length of hospital stay longer than 1 day were more frequently found in Group 1 (p = 0.024).
The median time length for readmission was 8.0 days with a range from 1 to 73 days. Of those with readmission, seven patients (18.9%) readmitted within the first 24 h after the discharge, and 11 patients (29.7%) readmitted between 1 and 7 days after the surgery. The distribution of the diagnoses seen in patients with readmission is displayed in table 4.
Diagnosis | ||||
---|---|---|---|---|
Surgical‡ | 39 (97.5) | |||
Biliary | 5 (12.8) | |||
Bile duct obstruction | 4 | |||
Bile duct injury | 1 | |||
Non-biliary | 34 (87.2) | |||
Post operative pain | 25 | |||
N/V | 5 | |||
Intestinal laceration | 1 | |||
Wound infection/hematoma | 3 | |||
Non-surgical‡ | 1 (2.5) |
‡n (%).
N/V: nausea/vomiting.
Surgical causes of both biliary and non-biliary etiology were detected in 97.5% of the cases. Symptomatic treatment without any intervention was applied to the most of the cases (91.9%). Surgical treatment and ERCP were necessitated in two and three patients. There was no mortality in patients with readmission.
Discussion
In this retrospective study, we detected that the rate of readmission following cholecystectomy was 6.16%. Obesity, the number of coexisting diseases, conversion to open surgery, the development of intraoperative complications, the use of drain, and length of hospital stay > 1 day were significantly associated with higher readmission rates.
Inpatient or ambulatory laparoscopic approaches have been included in the earlier studies evaluating the readmission rates after cholecystectomy3,4,6,8-12. However, these studies show significant differences regarding the inclusion criteria. Some authors included only laparoscopic approach and excluded the converted cases3,6,8. It has also been mentioned that conversion to open surgery is not an exclusion criterion, but there were no converted cases9,12. Only elective cases have been studied by some authors6,10. Besides, coexisting intraoperative cholangiography or laparoscopic bile duct exploration were also included in several studies4,11. Therefore, all these methodological differences should be considered during the evaluation of readmission rates after cholecystectomy. There were limited studies, in which all types of cholecystectomies, including laparoscopic, converted to open, and open through right subcostal incision were evaluated regarding readmission rate1,5. We, also, included all cholecystectomy approaches in the present study. Although the incidences of open and converted surgeries in these studies were low, we thought that admission rates after cholecystectomy might be generalized by using this approach.
In the previous meta-analyses and systematic reviews, the mean readmission rate was 3.3% ranging from 0% to 11.7%7,13. Other studies have reported similar readmission rates following cholecystectomy1,3,5,6,8-11. However, it should be kept in mind that the study groups in these studies have different demographic and clinical features. The period for these studies ranged from 30 days to 90 days1,3,4,8,11. We and others thought that a 30-day time period might underestimate the true incidence of complications and readmissions1,5,6,12. Although the readmission rates may increase as the period increases, we used a 90-day interval for the definition of readmission for a more comprehensive evaluation.
Post-operative pain, nausea, and vomiting were the most common causes of readmission besides surgical complications and infective diagnoses7,11,13,14. Such complications have been detected after both ambulatory and non-ambulatory laparoscopic cholecystectomy. McIntyre's meta-analysis7 showed that surgical complications, including post-operative nausea and vomiting, were seen in 76% of reported cases. In some studies, post-operative abdominal pain was detected in more than half of the cases who readmitted following laparoscopic cholecystectomy4,8. Although there has been no optimal multimodal analgesic treatment after laparoscopic cholecystectomy, these patients might receive inadequate pain relief11. In the present study, we did not have data about the medications that were prescribed and their use by the patients due to the retrospective nature of the study. The association between post-operative pain management and readmission rates remains controversial, and prospective studies are needed to clarify this issue.
The presence of vague symptoms such as pain, nausea, vomiting, hypotension, dehydration, ileus, and fever is a problematic issue for the physicians1,3,11. Besides, most of these symptoms occur during the post-operative 1st week3. Almost half of the diagnoses in the post-operative 1st week were pain, nausea, and vomiting in the present study. Besides, post-operative nausea and vomiting are general complaints seen in 20-30% of surgical patients after the application of general anesthesia for any reason15,16. Therefore, unplanned readmission rates after cholecystectomy may not be used as a measure of quality.
The causes of readmission varied according to the classifications used in the previous studies. Some authors also recommended to standardize criteria regarding the reasons for readmission5. For that purpose, Rana et al.9 classified the reasons for hospital readmissions following cholecystectomy as biliary and non-biliary surgical, and non-surgical causes. Regarding this classification, surgical biliary causes were detected in one-quarter of the cases5,9,10. However, there were only five cases (12.8%) with biliary complications in the present study. The variances in these subgroups may originate from the differences in the methodological properties of the studies and clinical features of the patients. Considering these differences, the causes of readmission in each study may not reflect the ideal situation. Thus, prospective standardized studies are needed to clarify controversial issues.
In the earlier studies, the authors found that there are some predictive risk factors for readmission. Older age, male sex, associated comorbidities, and emergency cholecystectomy have been considered significant risk factors for hospital readmission after cholecystectomy3,11.
Besides, the presence of comorbidities, acute cholecystitis, chronic steroid use, and the length of the operation longer than 2 h was other factors9. In the present study, we found that obesity, the number of coexisting disease, conversion to open surgery, and the development of intraoperative complications were the significant risk factors for readmission. We also showed that there was a significant association between the use of drain and readmission. The use of drain may be used to highlight the complexity of the case9.
Awolaran et al.4 reported that a longer stay in hospital decreased the readmission rate. This finding was not supported by the studies focusing on ambulatory laparoscopic cholecystectomy3,11. We, also, showed that the length of hospital stay >1 day was significantly associated with readmission. Hence, great attention can be considered when discharging the patients after the removal of in situ drains or the patients with longer hospital stays. Besides, it may be better to inform these patients about the increased risk of readmission.
Obesity is one of the most significant risk factors for post-operative morbidity and mortality17. Several authors questioned the impact of obesity on the outcomes of cholecystectomy. In Matteori's study17, day case laparoscopic cholecystectomy has had similar readmission rates in obese patients compared with non-obese. They classified the patients as obese if their body mass index is ≥ 30 kg/m2. Reeves et al.18 showed that only super obesity is associated with increased hospital readmission. In this study, there was a significant association between readmission and Grade II and Grade II obesity. Hence, we may conclude that as the grade of obesity increases, the risk for readmission after cholecystectomy also increases.
The retrospective nature of the study and the use of an administrative database were the major limitations. This type of study may cause loss of some potentially important variables such as discharge medications, and status of pain just before discharge due to the inherent nature of data. Although we aimed to include the readmitted cases with medical reasons, we could not find all medical reasons directly related to surgical procedures due to the lack of surgical consultations. We obtained data from the medical records of the patients using the hospital information system. Hence, difficulties during the evaluation of medical reasons for readmission were another limiting factor. Subjective assessment of pain in these patients was another limitation due to the lack of any pain score.
Conclusion
In conclusion, readmissions after cholecystectomy are rare conditions. Significant predictive factors may help physicians to be alerted during the discharge of the patients. Post-operative pain, nausea, and vomiting were the most common diagnoses which were relieved, usually by conservative management. Identifying the risky patients for readmission, developing discharge programs, and adequate post-operative pain relief can be regarded as essential steps for a significant reduction in readmission rates.