Discussion
To the authors’ knowledge, the herein presented study is the first to scrutinize differences in outcomes after minimally invasive valve surgery according to patient obesity class. Our results demonstrate mini-AVR to be feasible and safe in obese patients irrespective of obesity class, thus supporting a nondiscriminatory approach to operative planning for patients with valvular disease. The postoperative lengths of stay averaged four to five days and total direct costs $22,000-25,000, both outcomes comparable between patients of any obesity class. Thirty-day postoperative mortality affected two patients in the study population, and they were of Class I obesity, notably the largest patient cohort. Consequently, patients with comorbid obesity and aortic valve disease should be considered for minimally invasive surgery without concern for a disproportionately complicated course of recovery.
Performing minimally invasive valve surgery for obese patients has often been a point of contention due to concerns of adequate surgical field exposure, single lung ventilation, and the greater frequency of additional chronic medical conditions in this population. Nevertheless, studies comparing the outcomes of patients with normal (18.5-25 kg/m2), overweight (25-30 kg/m2), and obese (> 30 kg/m2) BMI have shown that experienced centers see comparable rates of postoperative mortality, reoperation, blood transfusion, wound infection, stroke, pacemaker implantation, and length of stay after minimally invasive valve surgery.9 Aljanadi et al. found that five-year survival between obese and non-obese patients undergoing minimally invasive mitral valve surgery was similar (95.8% vs 95.5%, p=0.83).10 In consideration of the patient with obesity, Santana et al. demonstrated minimally invasive approaches to be superior to full sternotomy with fewer postoperative complications (23.5% vs 51.0%, p=0.034) and lower in-hospital mortality (0% vs 8.3%, p=0.04).4
However, to date, these observations have not been scrutinized between obesity classes, raising question of the broad application of these findings across the continuum of higher BMIs. Our study thus builds on previous work to share multiple important findings. First, we demonstrated that mini-AVR is feasible in patients of Class I, II, and III obesity. Second, mini-AVR is safe despite baseline STS PROM scores that positively correlate with obesity class: patients amongst obesity classes experience no different lengths of stay, short-term mortality, hospital readmission rates, or perioperative complications (with the exception of atrial fibrillation). Third, the cost of performance of minimally invasive AVR (i.e. , via partial sternotomy or minithoracotomy) does not differ between patient obesity classes. Collectively, these observations support the equitable provision of minimally invasive options to obese patients regardless of obesity class – options that may in turn minimize perioperative pain, expedite recovery, and reduce resource utilization without compromising clinical outcomes.
Additional observations were made regarding the relationships between select baseline patient characteristics and obesity classes. Common to the current literature, we observed an association between obesity class and metabolic comorbidities, namely, diabetes mellitus;11 while our study found no difference in the frequency of diabetes between patients with Class I versus Class II obesity, diabetes was significantly more common amongst patients in the Class III obesity cohort. The STS PROM score also demonstrated a stepwise relationship with obesity class. In line with this finding, Ghanta and colleagues have shown that STS PROM tends toward lower risk scoring for obese patients (spanning Class I and II BMIs) versus normal weight, overweight, and morbidly obese (Class III) patients.12 In contrast to the current literature that correlates hypertension and obesity class, however, our study found no such association.12,13 This is due to an exceedingly high prevalence of hypertension across our study population (affecting 90-95% of patients on average), which would have required a large study population to differentiate minute differences amongst obesity cohorts.
Upon observation of the primary study outcomes, we found patient obesity class to have no effect on postoperative length of stay or direct cost of mini-AVR. Whilst Mariscalco et al. ’s obesity paradox in cardiac surgery continues to be debated, single center studies have shown that patients with higher BMI experience longer lengths of stay and more expensive costs of care after cardiac surgery.12,14 At our academic medical center, appropriate operative planning, intraoperative patient positioning, and, over the study period, a gradual implementation of typical Enhanced Recovery After Surgery (ERAS) protocol elements (e.g. , preoperative conditioning, early extubation) have perhaps prevented BMI from dictating operative room time and costly resource utilization. With a multidisciplinary team of cardiac surgery anesthesiologists and physician assistants, critical care physicians, and nurse practitioners providing comprehensive perioperative care, our patients are optimized for discharge within four to five days of their operation, as demonstrated in this study. Our study also failed to reveal a difference in 30-day mortality after mini-AVR amongst obesity classes, complementary to the meta-analysis conducted by Mariscalco et al. 3 Not only might obese patients experience paradoxically lower risks of mortality after cardiac surgery, but this risk may be comparable across all classes of obesity at least in the population of patient who undergo minimally invasive valve surgery.
Postoperative complications were generally experienced no differently by patients of any obesity classes. Only new atrial fibrillation was less frequently encountered by patients of Class I versus those of Class II and III obesity. With the incidence of onset of this arrhythmia as high as 20-50% after cardiac surgery, coupled with obesity as a known independent risk factor for atrial fibrillation, our finding should not preclude patients with Class II and II obesity from receiving minimally invasive valve surgery.15,16
There are several limitations to this study. First, bias can be attributed to its retrospective, observational design due to the lack of randomization, a control group, and a priori data field selection. Second, while our results demonstrated no difference in 30-day mortality amongst obesity classes, two patients in the study population did not survive the duration of the postoperative follow-up period. While we expect that these deaths occurred in the Class I obesity population by random chance consistent with its largest cohort size (total study population: Class I n=106 vs. Class II n=42 vs. Class III n=34), our study is under-powered to detect significant differences between such small event rates and thus to arrive at a sound conclusion regarding 30-day mortality amongst obesity classes. To better elucidate mortality differences amongst obese patients undergoing mini-AVR, a large database or multi-center study should be conducted.