Page 35 - Read Online

Page 4 of 7 Kheirvari et al. Mini-invasive Surg 2020;4:3 I http://dx.doi.org/10.20517/2574-1225.2019.45

Table 1. The distribution of study variables
Variables Total (n = 199) Case (n = 50) Control (n = 149)
Sex (count/percent)
Male 60 (30.2%) 15 (30%) 45 (30.2%)
Female 139 (69.8%) 35 (70%) 104 (69.8%)
Age (mean) 38.15 37.32 38.42
ESR (mean) 41.83 73.1 31.34
ESR: Erythrocyte sedimentation rate

Statistical analysis
Adjusted and unadjusted binary logistic regression models were used to evaluate effects of independent
variables on leaking outcome (0 = no, 1 = yes). Independent variables included sex, age, and ESR. The
significance level was defined as 0.05 (a = 0.05). Both adjusted and unadjusted variables with significant
levels were included in the final models and are reported below. Statistical analysis was performed using
IBM SPSS Statistics 25 (SPSS Inc., Chicago, IL).

The final predicting model for leaking outcome was designed using the following regression model:
log[P / (1 - P )] = a + b X
x
1 1
x
The final adjusted prediction model of log (odds) for leaking outcome was calculated using the following
equation:
y = (-3.576) + 0.50 (ESR)

RESULTS
We investigated, among the 2,350 patients who underwent sleeve gastrectomy from 2016 to 2019, 50
subjects who experienced gastric leak (2.12%). The total sample size was 199 patients, including 50 cases
experiencing leak and 149 controls (randomized from 2,350 patients). Overall, 69.8% of the cohort, 70%
of cases and 69.8% of controls, were females. The mean age for the cohort was 38.15 (minimum 12 years
old and maximum 63 years old). The mean ESR was 73.1 mm/h for cases, which is statistically significantly
higher in patients with leakage compared to the control group. In addition, ESR serum level mean was
31.34 mm/h for control groups. More descriptive results are reported in Table 1.

The results of adjusted and unadjusted logistic regression are reported in Table 2. Females were taken
as reference group due to bigger proportion in the sample. The only independent variable which had
significant association with staple-line leakage was ESR (OR = 1.051). This means that, for every 1 unit
increase of ESR, the odds for staple-line leakage occurrence increases by 5.1%.

DISCUSSION
The sleeve gastrectomy procedure has been popularly employed for the management of morbid obesity and
this operation has a series of inevitable complications. Staple-line leakage is one of these complications,
[6]
with an incidence ranging from 7% to 25% after bariatric surgery . Although researchers have mentioned
various approaches, surgeons utilize the endoscopic approaches, such as stent inserting, clips, and biologic
[16]
glue . This study comprised our experiences with 199 patients, with or without staple-line leakage,
after sleeve gastrectomy. The final adjusted prediction model of log (odds) for leaking outcome can be
used to predict leaking outcome. Exponential of (y) gives odds of occurrence of leaking for each patient
with archived ESR. It should be noted that, because of the impact of other factors affecting on staple-line
leakage outcome, this model might not be 100% precise.