Risk factors for patient selection in ambulatory laparoscopic cholecystectomy: A Single-centre experience.

Backgrounds: Ambulatory laparoscopic cholecystectomy (LC) reduces healthcare cost and increases hospital bed capacity. Currently, there is no consensus on patient selection for ambulatory LC. Evaluation of risk factors for ambulatory discharge is essential. Materials and
Methods: Consecutive patients who underwent LC in our centre throughout 2019 were collected. We evaluated the discharge fitness using the Post-Anaesthetic Discharge Scoring System at 8 h after the operation. The relations between pre-operative variables and dischargeable possibilities were analysed for screening risk factors. Furthermore, we performed a literature review to summarise all published information.
Results: Six hundred and forty-one cases were included in this study. American Society of Anaesthesiologist (ASA) grading (odds ratio OR = 0.415, P = 0.001) and leucocytes (OR = 0.80, P < 0.001) significantly predicted the fitness of discharge. ASA contributed to lower activity (P = 0.002) and intake/output (P = 0.006) scores. Leucocytes influence the vital sign (P < 0.001) and pain or post-operative nausea and vomiting (PONV) (P < 0.001) scores. The prolonged operation could predict the inabilities of discharge with a cut-off value of 55 min by dropping vital signs (P = 0.011), activity (P < 0.001) and pain or PONV (P = 0.012) scores. Male sex (OR: 1.702, P = 0.010), body mass index (BMI) (OR: 1.087, P = 0.008), leucocytes (OR: 1.075, P = 0.017) and C-reactive protein (CRP) (OR: 1.018, P = 0.003) were predictors for prolonged operation (>55 min). Conclusions: We suggest that pre-operative ASA grading III and leucocytes are risk factors for the fitness of ambulatory discharge after LC and intraoperative time. Male, BMI and CRP predict complicated surgery, and they should be considered preoperatively.

INTRODUCTION

Ambulatory laparoscopic cholecystectomy (LC), also known as day-case LC or outpatient LC, could reduce healthcare cost and increase hospital bed capacity.[12] However, this procedure has not been widely acknowledged, especially in developing countries. Many surgeons’ main hesitation about this procedure is the possible misdetection of the appearance of any vital post-operative complications. Therefore, basic principles are necessary for determining the ambulatory process and ensuring the highest probability of success with the utmost safety for candidate patients.

In general, the patients should be selected based on three individual aspects: social, medical and surgical factors.[3] Social requirements are fully understanding and consent, a responsible adult carer and convenient transportation. The most considered medical factors are the American Society of Anaesthesiologist (ASA) physic status, age and body mass index (BMI). However, there is not enough evidence for optimal cut-off values for each parameter. Some studies also reported the satisfactory outcome of ambulatory LC in those patients with high ASA score, elder age, or obesity.[45] In the surgical aspect, the patients with a high risk of severe post-operative complications should be excluded. In previously published studies, some researchers excluded the patients with acute cholecystitis, history of pancreatitis or abdominal surgery or clinical suspicion of common bile duct stones.[678910111213] In contrast, some other studies also reported acceptable outcomes with those patients.[1415161718] Besides, abnormal laboratory tests and thickening gallbladder wall in ultrasonography have been well known to be risk factors predicting difficult operation,[19202122] but only a few studies considered them as patient's selection criteria.[232425]

To develop a suitable selection criterion for further implementing the ambulatory procedure in our centre, we conducted this observational study of unselected consecutive LCs, and we investigated the risk factors for the satisfaction of ambulatory discharge criteria.

MATERIALS AND METHODS

Patients

We collected all the consecutive patients who underwent LC from January 1 to December 31, 2019, in our single centre. This study's exclusion criteria were (1) children, pregnancy and the disabled and (2) confirmed diagnosis of choledocholithiasis. Due to the limited acceptance by local culture, ambulatory LC has not been carried out regularly in our centre. We admitted all the patients to the inpatient ward and routinely discharged patients 2 days after surgery. No effort was made to avoid complicated and difficult patients. All patients were written informed consent with this observational study. This study was approved by the ethical review committee of The First Affiliated Hospital of Chengdu Medical College. The Declaration of Helsinki was strictly followed during all procedures of this study.

Pre-operative management

Ultrasonography was routinely adopted for diagnosing gallbladder disease and measuring gallbladder wall thickness in all patients preoperatively. The diagnosis of ongoing acute cholecystitis was made according to the Tokyo Guidelines 2018.[19] Chronic cholecystitis was defined as recurrent upper abdominal discomfort plus the imaging evidence of gallstones or polyps. Laboratory tests, including a blood test, liver and renal function and coagulation, were examined routinely and recorded. Furthermore, all the patients were preoperatively instructed to follow the enhanced recovery after surgery basis during hospitalization, involving detailed pre-operative counselling, short fasting duration, avoidance of prophylactic nasogastric tubes and early mobilization. ASA physical status classification was evaluated by an anaesthesiologist preoperatively according to the 2014th edition.[26]

Operation procedure

The operations were performed by a consultant surgeon and an attending surgeon with a three-port technique under general anaesthesia. Prophylactic abdominal drainage was placed only in cases of the difficult procedure with a high risk of post-operative bleeding or bile leakage. Injection with ropivacaine hydrochloride (10 mg/ml) on all port sites was performed at the end of the operation.

Post-operative management

After the operation, patients were encouraged to mobilise as soon as possible and offered oral fluids as tolerated. We measured the suitability for discharge of all patients at approximately 8 h after operation by utilizing the Post-Anaesthetic Discharge Scoring System (PADSS).[27] [Table 1] however, as mentioned above, we kept those patients meeting the criteria instead of really discharging them for safety and ethics concerns. The reasons for unsatisfying the criteria were recorded and analysed. All patients without any complications were usually discharged 48 h after the operation.

Table 1

Post-Anaesthetic Discharge Scoring System used in the current study

Score	Evaluation
Vital signs
2	Within 20% of the pre-operative value
1	20%-40% of the pre-operative value
0	>40% of pre-operative value
Activity and mental status
2	Oriented x3 AND has a steady gait
1	Oriented x3 OR has a steady gait
0	Neither
Pain or PONV
2	Minimal
1	Moderate, having required treatment
0	Severe, requiring treatment
Surgical bleeding
2	Minimal
1	Moderate
0	Severe
Intake/output
2	Has had oral administration of fluids and voided
1	Has had oral administration of fluids OR voided
0	Neither

A summated score of 9-10 indicates the fitness for discharge. PONV: Post-operative nausea and vomiting. Orientation is often assessed as part of a mental status test. "Oriented x3" means awareness of PERSON, PLACE and TIME. In addition, x1: only PERSON, x2: only PERSON and PLACE, x4: PERSON, PLACE, TIME, and SITUATION

Statistical analysis

Comparison between two grouped continuous data was conducted by using the independent t-test (normally distributed) or the Mann–Whitney u test (non-normally distributed), and assessment among three or more grouped data was analysed by utilizing the one-way ANOVA (normally distributed) or the Kruskal–Wallis test (non-normally distributed). The Chi-square test or Fisher's exact test was employed for comparisons of contingency variables. The significance of operation time on dischargeable possibility was examined by receiver operating characteristic (ROC) analysis. The optimised cut-off value was selected by the highest Youden index (sensitivity%+specificity%-1). Univariate analysis was conducted for screening potential variables. Those variables with a level of significance of ≤0.100 were then included in the multivariate analysis by stepwise logistic regression. The scores of each section of PADSS from all patients were further analysed. All statistical procedures were performed with Prism (Version 8.4.0, GraphPad Software, USA) or SPSS (Version 26.0.0.1, IBM, USA) statistics software. P < 0.050 was considered statistically significant.

RESULTS

American Society of Anaesthesiologist grading, leucocyte counts and direct bilirubin predicted the fitness of ambulatory discharge

In the current study, we eventually included 641 consecutive patients who underwent LCs. Eight (1.3%) patients had conversions to open surgery intraoperatively. Four hundred and ten (64.0%) patients were female. The median age was 50.54 (±23.3) years and the median BMI was 24.52 (±4.35). Most cases were within ASA Grade I and II; 73 (11.4%) patients had Grade III and no patients had Grade IV or higher.

In univariate analysis, as shown in Table 2, in the non-dischargeable group, there were more ASA III patients (P < 0.001) and higher leucocyte counts (P < 0.001) and C-reactive protein (CRP) (P = 0.035). We also observed the tendency of higher levels of gamma-glutamyl transferase (GGT) (P = 0.100) and direct bilirubin (P = 0.070) in the non-dischargeable group. These five variables were then included for the following multivariate analysis.

Table 2

Demographics of all included patients and univariate analysis of pre-operative variables between dischargeable and non-dischargeable patients

Variable	All patients (n=641), n (%)	Dischargeable (n=314), n (%)	Non-dischargeable (n=327), n (%)	P
Sex (male/female)	231/410	117/197	114/213	0.565
Age	50.21 (±23.10)	50.48 (±20.10)	49.83 (±24.50)	0.495
BMI	24.52 (±4.35)	24.43 (±4.34)	24.63 (±4.44)	0.235
ASA III	73 (11.4)	20 (6.4)	53 (16.2)	<0.001
Cholelithiasis	616 (96.1)	299 (95.2)	317 (96.9)	0.310
Cholecystitis
Acute	362 (56.5)	167 (53.2)	195 (59.6)	0.158
Chronic	219 (34.2)	112 (35.7)	107 (32.7)
Silent	60 (9.4)	35 (11.1)	25 (7.6)
History of upper abdominal surgery	20 (3.1)	9 (2.9)	11 (3.4)	0.822
History of pancreatitis	29 (4.5)	12 (3.8)	17 (5.2)	0.451
Thickening gallbladder wall	305 (47.6)	142 (45.2)	163 (49.8)	0.425
Leucocytes (109/L)	8.86 (±4.48)	8.29 (±3.72)	9.46 (±5.02)	<0.001
ALT (U/L)	25.00 (±29.00)	26.00 (±24.25)	25.00 (±31.00)	0.923
AST (U/L)	23.00 (±13.00)	22.00 (±11.00)	23.00 (±15.00)	0.485
GGT (U/L)	31.00 (±50.00)	29.00 (±47.00)	32.00 (±57.00)	0.100
ALP (U/L)	97.00 (±41.50)	97.00 (±40.25)	97.00 (±44.00)	0.521
Albumin (g/L)	43.90 (±5.90)	44.30 (±5.33)	43.50 (±6.50)	0.147
Direct bilirubin ( mol/L)	3.80 (±2.70)	3.60 (±2.70)	3.90 (±2.70)	0.070
CRP (mg/L)	2.00 (±4.00)	2.00 (±5.00)	1.30 (±3.00)	0.035

All continuous data above were abnormally distributed. They were examined by Mann-Whitney U-test. Categorised data were analysed by Pearson Chi-square tests. BMI: Body mass index, ASA: American Society of Anaesthesiologist, ALT: Alanine transaminase, AST: Aspartate transaminase, GGT: Gamma-glutamyl transferase, ALP: Alkaline phosphatase, CRP: C-reactive protein

After the stepwise binary logistic regression process, GGT, direct bilirubin and CRP were excluded. Eventually, the ASA III (P = 0.012, odds ratio [OR] = 0.415, 95% confidence interval [CI]: 0.209–0.821) and leucocytes (P < 0.001, OR = 0.870, 95%CI: 0.821–0.922) were significant predictors for the satisfaction of discharge criteria.

Higher American Society of Anaesthesiologist grades predicted lower activity and intake/output scores

The relationships of ASA grading and leucocytes with sectional scores from PADSS were further analysed. It shows that the ASA III patients had significantly lower activity scores (P = 0.002) and lower intake/output scores (P = 0.006). No significance was observed between ASA grading and PADSS vital sign (P = 0.142), pain or post-operative nausea and vomiting (PONV) (P = 0.291) or surgical bleeding (P = 0.517) scores.

Higher leucocyte counts predicted lower vital sign and pain or post-operative nausea and vomiting scores

As shown in Figure 1, the results showed that the patients with a lower post-operative vital sign [P < 0.001, Figure 1a] and pain or PONV scores [P < 0.001, Figure 1b] had significant pre-operative leucocyte counts. There is no significance between leucocyte counts and PADSS activities (P = 0.234), surgical bleeding (P = 0.760) or intake/output (P = 0.102) scores.

Figure 1

The Post-Anaesthetic Discharge Scoring System vital sign (2: within 20% of pre-operative value; 1: 20%–40% of pre-operative value; 0: >40% of pre-operative value) (a) and Pain or post-operative nausea or/and vomiting (2: Minimal; 1: Moderate, having required treatment; 0: Severe, requiring treatment) (b) scores were significantly related to pre-operative leucocyte counts (presented as median value), significance was examined by Kruskal–Wallis test

Operation time predicted the fitness of discharge

The operation time was also evaluated for the fitness of discharge. In total, the median operation time was 57.00 (±32.00) min. The non-dischargeable patients had significantly longer operations than the dischargeable (60.00 [±44.00] vs. 53.50 [±24.00], P < 0.001). ROC analysis also revealed that operation time could predict failure for discharge [Figure 2, area under curve = 0.611, P < 0.001). The most predictive cut-off value with the highest Youden index was 55 min (sensitivity% = 63.6%, specificity% = 56.7%, Youden Index = 0.203).

Figure 2

Receiver operating characteristic analysis for the predictive effectiveness of operation time on the fitness of ambulatory discharge. AUC, the area under the curve

We continuously analysed the relations between operation time and PADSS sectional scores. The results showed that those patients with lower vital signs [P = 0.011, Figure 3a], activity [P < 0.001, Figure 3b] and pain or PONV [P = 0.012, Figure 3c] scores had significant prolonged operations. No significance was found for surgical bleeding (P = 0.116) or intake/output (P = 0.138) scores.

Figure 3

The Post-Anaesthetic Discharge Scoring System vital sign (2: within 20% of pre-operative value; 1: 20%–40% of pre-operative value; 0: >40% of pre-operative value) (a), Activity (2: Oriented*3 AND has a steady gait; 1: Oriented*3 OR has a steady gait; 0: Neither) (b), and Pain or post-operative nausea or/and vomiting (2: Minimal; 1: Moderate, having required treatment; 0: Severe, requiring treatment) (c) scores were significantly related to operation time (presented as median value), significance was examined by Kruskal–Wallis test

Male patients, higher body mass index, leucocytes and C-reactive protein predicted prolonged operation

We further conducted univariate and multivariate analyses of predictive variables for prolonged operation (55 min calculated by ROC grouped patients). It showed that male patients (P < 0.001), acute cholecystitis (P < 0.002), presence of thickening gallbladder wall (P = 0.033), higher ASA grading (P = 0.046), BMI (P < 0.001), leucocytes (P < 0.001), GGT (P < 0.001), alkaline phosphatase (P = 0.015), albumin (P < 0.026), direct bilirubin (P = 0.008) and CRP (P < 0.001) were significantly related to prolonged operation time. After stepwise logistic multivariate analysis, male patients (P = 0.010, OR = 1.702, 95% CI: 1.134–2.554), higher BMI (P = 0.008, OR = 1.087, 95% CI: 1.022–1.156), leucocytes (P = 0.017, OR = 1.075, 95% CI: 1.013–1.140) and CRP (P = 0.003, OR = 1.018, 95% CI: 1.006–1.029) were independent predictors for prolonged operation.

DISCUSSION

In the present study, we discovered the influence of pre-operative ASA grading and leucocytes and operation time on ambulatory discharge criteria’ satisfaction. We also found the significant risk factors, which might indirectly affect discharge's fitness, for prolonged operation. As well, we revealed the reasonings for failure discharge behind these influences.

As per the suggested guidelines from the British Association of Day Surgery in 2019, the patient's selection criteria should fall into three major aspects: social, medical and surgical.[3] Regarding the medical aspect, our study proved that higher ASA grading indicated restricted post-operative activity and digestive function, resulting in failure for ambulatory discharge. We did not observe the significance of age in the current study, and we suggest that the ambulatory LCs are acceptable in the elderly with a stable medical condition. This is in accordance with a previous research based on a large cohort from a national database.[28] Meanwhile, there was no statistically significant difference for BMI in our results. A previous reported similar result from a European research group suggested that ambulatory LC was safe in obese patients with similar admission rates to non-obese controls.[4] However, the influence of BMI on the prolonged operation was indeed proved in the current study. In Asian countries, extreme obesity is not very common.[29] In our cohort, there were only two patients who had a BMI of over 35. In fact, in the latest ASA classification version in 2014, BMI has been considered during ASA evaluation by anaesthesiologists.[26] Therefore, we recommend that ASA grading is effective for pre-operative assessment, and the patients with ASA III should be excluded from ambulatory LC.

Leucocyte counts could significantly reflect complicated operations and unsatisfied dischargeable rates in this study. Many previous publications have suggested the leucocyte count as a predictor for difficult LC.[193031] However, only one previous study directly employed leucocytes in selection criteria for ambulatory LC.[23] Besides leucocyte counts, we confirmed the influence of CRP on operation time. Similar results for the predictive role of CRP on difficult operations were also reported in earlier publications.[2232] Our study noticed the possible impacts of acute cholecystitis and other inflammatory markers, such as thickening gallbladder wall, alanine aminotransferase, GGT, albumin and direct bilirubin on difficult surgery in univariate analysis. However, they were all excluded by multivariate analysis according to the interactions with leucocytes and CRP. We suggested that in comparison with using the type of diagnosis, the objective and measurable leucocyte is more valuable to evaluate the operational difficulties and predict the fitness of ambulatory discharge.

Another independent risk factor for difficult operation is the male sex. The increased difficulty for men in LC has been widely reported in several large cohort studies.[333435] However, the reason remains unclear, and speculations were lower compliance of medical advice, higher pain threshold and more intra-abdominal fat distribution in men.

The limitation of the current study was not an experience from the actual practice of ambulatory LC in ignorance of social factors, which are also important. Furthermore, we had a limited number of traditionally considered high-risk patients, such as those with extreme obesity or high age. This may cause insufficiency in statistics.

CONCLUSIONS

Our study suggests that pre-operative ASA grading III and leucocytes are risk factors for fitness of ambulatory discharge after LC, as well as intraoperative time. Male, BMI and CRP predict complicated surgery, and they should be considered preoperatively.

Financial support and sponsorship

This research was granted by the Institutional Foundation of the First Affiliated Hospital of Chengdu Medical College (CYFY-GQ20).

Conflicts of interest

There are no conflicts of interest.