Is Laparoscopic Bariatric Surgery a Safe Option in Extremely High-Risk Morbidly Obese Patients?

Introduction

Bariatric surgery is the cornerstone of any effective long-term treatment for morbid obesity and its associated medical comorbidities such as type 2 diabetes, hypertension, dyslipidemia, obesity hypoventilation syndrome, and obstructive sleep apnea.^1–3 Consequently, the positive impact on the metabolic syndrome is assumed to be responsible for the reduction in long-term mortality associated with morbid obesity.^4–6 Nevertheless, the surgical treatment of the morbidly obese with significant medical conditions remains a challenge.

Studies have demonstrated the link between certain patient factors, especially significant comorbidities, with an increase in morbidity and mortality associated with bariatric surgery, particularly in the superobese population (body mass index [BMI] of ≥50 kg/m²).^7,8 The center-dependent comparison of surgical outcomes has also been made in several population-level studies.^9,10 However, there are only a few validated risk stratification indices that estimate the likelihood of negative postoperative outcomes based on the baseline status of the patients.^11–13

The Obesity Surgery Mortality Risk Score (OS-MRS) is the most widely used index, which identifies a high-risk group of patients based on sex, age ≥45 years, superobesity, hypertension, and history of a venous thromboembolic event. ¹³ The postoperative mortality risk based on OS-MRS is estimated at 7.56% for those high-risk patients, which may deter referrals for this group of patients who could arguably benefit greatly from bariatric surgery.

However, commonly used risk calculators are based on data that include patients who had an open bariatric procedure, which is associated with more postoperative complications. Hence, risk scores may not be fully reflective of the current expected surgical outcomes. High volume and near-ubiquitous utilization of a minimally invasive approach, combined with advancement in perioperative care, have further mitigated postoperative morbidity in such high-risk patients. Bariatric surgery candidates usually have an extensive range of potential medical, surgical, and psychological comorbidities. These individuals should be carefully selected, extensively evaluated, and optimized in order to achieve best outcomes following an elective surgery. Therefore, a comprehensive multidisciplinary preoperative assessment is of great importance, especially in high-risk patients. In this study, we present our outcomes in extremely high-risk patients who underwent laparoscopic bariatric surgery.

Materials and Methods

Between January 2006 and June 2012, in total, 3240 patients underwent laparoscopic bariatric surgery by five bariatric surgeons at a single academic institution. We constructed our study cohort by expanding on the criteria used for identifying high-risk patients as originally described for the OS-MRS. ¹³ We strictly defined an extremely high-risk group of patients by age at the time of surgery of ≥65 years, BMI of ≥50 kg/m², and presence of at least two of six cardiopulmonary or vascular comorbidities (hypertension, ischemic heart disease, congestive heart failure, chronic obstructive pulmonary disease, obstructive sleep apnea, and history of deep vein thrombosis or pulmonary embolism).

The following patient- and surgery-related variables were obtained from the electronic hospital patient charts: age, gender, baseline BMI, American Society of Anesthesiologists class, cardiopulmonary comorbidities, and operative time. Subsequently, the following outcome variables were assessed: length of hospital stay, 30-day postoperative morbidity, mortality, re-operation, re-admissions, delayed complications, and weight loss. Percentage excess weight loss was calculated as follows: ([operative weight – follow-up weight]/[operative weight – ideal weight])×100. Ideal weight was based on a BMI of 25 kg/m². Percentage total weight loss was defined as follows: ([operative weight – follow-up weight]/operative weight)×100. Definitions of improvement of hypertension and diabetes following bariatric surgery have been presented elsewhere. ³ In brief, diabetes remission was defined as a normal fasting glucose and a glycated hemoglobin level of <6.5% off medication. Hypertension remission was defined as normal blood pressure measurements off medication.

Follow-up was achieved through medical records using data from the follow-up clinic visits by the surgeon, bariatrician, nurse, or the dietitian. Follow-up visits were scheduled at 1–2 week(s), 1 month, 3, 6, 9, 12, and 18 months, and yearly thereafter following the operation.

The Institutional Review Board approved the study proposal and access to patients' charts.

Continuous variables with normal distribution, variables with non-normal distribution, and categorical variables were reported as mean±standard deviation, median (range), and frequencies (%), respectively. Statistical analysis was carried out using Student's t test and Fisher's exact test in STATA software (Stata Corp, College Station, TX) version 12. Inference was based on a two-sided 5% level.

Results

Of 3240 patients who underwent laparoscopic bariatric surgery at a single institution between January 2006 and June 2012, based on our definition, 44 patients (1.4% of all patients) were considered to be extremely high risk. Patients had a male-to-female ratio of 9:35, a mean age of 67.9±2.7 (range, 65–76) years, and a mean BMI of 54.8±5.5 (range, 50–82) kg/m². The median number of cardiopulmonary and all comorbidities were two (range, two to five) and seven (range, 5–13), respectively. Twenty-seven patients had type 2 diabetes mellitus. Eight patients had American Society of Anesthesiologists class IV physical status, and 36 patients had class III. The baseline characteristics of the study cohort are shown in Table 1.

Laparoscopic surgical procedures included Roux-en-Y gastric bypass (RYGB) (n=23), adjustable gastric banding (AGB) (n=11), and sleeve gastrectomy (SG) (n=10). One operation was a revisional bariatric surgery (AGB to RYGB), and a simultaneous ventral hernia repair was performed in 4 cases. There were no intraoperative complications or a conversion to open technique. The mean operative time and length of stay were 170.9±56.9 minutes and 4.1±3.0 days, respectively, and varied as expected according to procedure type.

Thirteen (29.5%) 30-day postoperative complications occurred. Among them, six (13.6%) complications were major, including a leak of the gastrojejunal anastomosis, adhesive obstruction, pulmonary embolism, myocardial infarction, atrial fibrillation, and respiratory failure. All of the major early complications except one (n=5) occurred after RYGB; however, the difference among surgical groups did not reach the statistical significance (P=.19). The 30-day postoperative re-admission, re-operation, and mortality rates were 15.9%, 2.3%, and 0%, respectively (Table 2).

Table 3 summarizes the intermediate-term outcomes following the bariatric surgery with a mean follow-up time of 24.0±18.4 months. Nine patients (20.4%) had a postoperative follow-up of<1 year. The mean percentage total weight and excess weight losses after at least 1 year of follow-up were 26.7±12.0% and 44.1±20.6%, respectively. The late morbidity and mortality rates during the length of the follow-up were 18.2% and 2.3%, respectively. Remission or improvement of diabetes and hypertension was observed in 71.4% of diabetic and 74.3% of hypertensive patients, respectively.

Statistical analysis revealed no significant correlation between age, BMI, or number of cardiopulmonary or total comorbidities with either perioperative or late adverse outcomes in this cohort of high-risk patients (data not shown).

Discussion

The morbidity and mortality profile associated with bariatric procedures has significantly improved in the last two decades.^14,15 Even though primary laparoscopic bariatric procedures are generally considered safe with a 30-day mortality rate of less than 0.3% following AGB, SG, or RYGB,^1,8 the risk of postoperative death increases as the patient's baseline risk factors increase. DeMaria et al. ¹³ developed the widely used OS-MRS with which they identified a high-risk group of patients that included superobese hypertensive males who were 45 years or older with a pulmonary embolus risk. The 90-day operative mortality rate for this high-risk group was 7.6%. ¹³ However, we report a 0% 30-day mortality rate in a group of extremely high-risk patients after bariatric surgery. Our study cohort was selected using a stricter set of criteria compared with the OS-MRS, including older (≥65 years old) superobese patients who had at least two of six cardiopulmonary comorbidities, including hypertension, ischemic heart disease, congestive heart failure, chronic obstructive pulmonary disease, obstructive sleep apnea, and history of deep vein thrombosis or pulmonary embolism. It should be noted that the primary intention of this study was to assess the safety of bariatric surgery in a cohort of very high-risk patients, not to introduce a new risk model for adverse events.

Furthermore, the long-term mortality (cause unknown) of this extremely high-risk group was only 2.3% (1 patient) during the mean follow-up time of 2 years, which remains significantly lower than the 7.6% 90-day mortality reported by DeMaria et al. ¹³ This difference in both early and late mortality rates could be partly explained by including open RYGB cases in generating the OS-MRS. In fact, Sarela et al. ¹⁶ demonstrated that, if the operation is done laparoscopically, the difference in operative mortality disappears for the different OS-MRS risk groups. Moreover, the type of surgery is independently predictive of adverse outcomes, with an increasing order of mortality associated with AGB, SG, and RYGB, respectively. Our study was not designed to answer the question of what surgery is best for the high-risk patients. We, however, report six major complications that occurred in our cohort (13.6%), and all but one occurred after RYGB. The difference among surgical groups did not reach statistical significance due to a small sample size. Thus the type of surgery is important in this group of older superobese patients. Perhaps a durable procedure with less morbidity and mortality such as SG is more appropriate and can also be devised into a two-step surgical approach, which could reduce the early operative mortality and improve the comorbidity profile in this group of patients.

Good outcomes in this patient population depend on several factors. The first is the use of minimally invasive techniques in performing bariatric procedures. ¹⁷ In our study, we excluded patients who had open bariatric procedures to assess the utility of laparoscopic surgery in reducing the operative risk for the high-risk patients. This point is demonstrated in a study by Nguyen et al. ¹⁸ using the National Inpatient Sample database to examine factors associated with mortality in 304,515 patients who underwent bariatric surgery in the period from 2006 to 2008. Open bariatric surgery was found to be the only modifiable risk factor impacting the early postoperative mortality, and the remaining predictors were all deemed to be nonmodifiable. ¹⁸

The second major determinant may be the performance of bariatric surgery at high-volume centers for this type of patient. In another study by Nguyen et al., ¹⁹ the authors evaluated data on all patients who underwent RYGB from 22 high-volume, 27 medium-volume, and 44 low-volume centers using the University Health System Consortium Clinical Database. They found lower morbidity and mortality at the high-volume centers, where the observed mortality was four times lower than the other centers. When analysis was carried out in a subset of patients older than 55 years, the observed mortality was 0.9% at high-volume centers versus 3.1% at low-volume centers. ¹⁹ This difference in mortality may in part be attributable to the early recognition and rescue of patients who have complications in high-volume centers.

The third factor is the proper identification of the high-risk patients preoperatively and the implementation of a multidisciplinary team to optimize the patient. At the Cleveland Clinic in Ohio, we have developed different preoperative evaluation pathways, tailored to specific groups of potential bariatric patients. The details of how patients are triaged and the components of each pathway have been presented elsewhere. ²⁰ In brief, all patients deemed to be medically high risk initially undergo a thorough medical assessment by one of three internists who focus their practice on the bariatric patients. Based on this assessment, the patient is referred for additional investigations or consultation with other specialties as indicated. Preoperatively, all patients are also seen by a nutritionist and a psychologist specializing in bariatric patients. The respective services follow the patients postoperatively as well. We believe thorough risk stratification and risk factor optimization as well as postoperative medical expertise and management for these high-risk patients are essential.

The retrospective nature of assessment, the small sample size of this cohort, and incomplete postsurgical follow-up are limitations of our study. However, our results represent a consistent theme with other studies that have evaluated bariatric surgery in the high-risk population, especially when proper risk-reduction measures were implemented.^17,20–22 In addition, our study reported the operative outcomes during an intermediate follow-up period of 2 years that demonstrate the metabolic and weight loss benefits of surgery in these patients. Additional studies with a focus on the long-term outcomes of bariatric surgery in such high-risk patients are needed.

In conclusion, we have demonstrated that laparoscopic bariatric surgery in extremely high-risk patients can be done safely with acceptable early and late morbidity and mortality rates. Advanced age, high BMI, and severe cardiopulmonary comorbidities should not exclude patients from consideration for bariatric surgery.