The Burden of Obesity in Cardiac Surgery: A 14 years' Follow-Up of 14.754 Patients

Aims The prevalence of obesity is rapidly increasing during the past decades. While previous research has focused on the early outcome after cardiac surgery or specific complications, the current study covers the whole burden of obesity in the field of cardiac surgery over short term and long term. Endpoints of the study were all-cause mortality, perioperative outcome, and wound-healing disorders (WHDs). Methods 14.754 consecutive patients who underwent cardiac surgery over a 14 years' time period were analyzed. BMI classifications were used according to the WHO definition. Results Mean survival was 11.95 years ± 0.1; CI 95% [12.04–12.14]. After adjustment for clinical baseline characteristics, obesity classes' I–III (obesity) did not affect 30-day mortality or all-cause mortality during the whole observational period. After adjustment for known risk factors, the risk for WHDs doubled at least in obesity patients as follows: obesity I (OR = 2.06; CI 95% [1.7–2.5]; p < 0.0001), obesity II (OR = 2.5; CI 95% [1.83–3.41]; p < 0.0001), and obesity III (OR = 4.12; CI 95% [2.52–6.74]; p < 0.0001). The same applies to the risk for sternal reconstruction that is substantially elevated in obesity I (OR = 2.23; CI 95% [1.75–2.83]; p < 0.0001), obesity II (OR = 2.81; CI 95% [1.91–4.13]; p < 0.0001), and obesity III (OR = 2.31; CI 95% [1.08–4.97]; p=0.03). No significant correlation could be found between obesity and major adverse events in the perioperative course like renal failure, ventilation >24 h, re-exploration, or cerebrovascular events. Conclusions Cardiac surgery is safe in obesity as short- and long-term mortality are not increased, and major adverse events during the perioperative course are similar to control patients. The burden of obesity lies in substantially increased rates of wound-healing disorders and sternal reconstructions.


Introduction
According to the World Health Organization (WHO), the worldwide prevalence of obesity nearly tripled between 1975 and 2016.In 2016, nearly 39% of men over 18 years were overweight with a Body Mass Index (BMI) of 25-30 and 13% were obese with a BMI >30 (the WHO European Regional Obesity Report 2022, Copenhagen).Regarding the worldwide trends in underweight and obesity from 1990 to 2022, the combined burden of underweight and obesity has increased in most countries, driven by an increase in obesity [1].Obesity is a well-established risk factor for multiple cardiovascular diseases, such as coronary disease, coronary death, and congestive heart failure [2].It increases the likelihood of occurrence and severity of cardiovascular risk factors, including dyslipidemia, diabetes, hypertension, and sleep disorders [2,3].Te increasing prevalence of obesity makes it one of the most critical public health problems worldwide, with enormous implications for treatment strategies and costs.
As obese patients are at higher risks for cardiovascular diseases, consequently the number of obese patients needing cardiac surgery will increase.Previous studies showed that the impact of obesity on perioperative mortality and major in-hospital adverse events as well as on early outcome after cardiac surgery is variable [4][5][6][7][8][9][10].Variable results are also reported for subgroups like valve surgeries [5,11] or patients with acute type A aortic dissections [12,13].In contrast, extreme obesity with a BMI >40 had signifcant increase in length of stay, rate of renal failure with necessity of renal replacement therapy, or prolonged ventilation compared to nonobese patients [14].No diference was found in the rate of stroke [13][14][15][16][17][18][19].Tese fndings may lead to increased riskadjusted hospital costs that might be up to 17% higher in obese patients undergoing cardiac surgery [9].
Studies analyzing the long-term outcome are lacking, while previous research has mainly focused on the early outcome after cardiac surgery.While previous research has focused on the early outcome after cardiac surgery or specifc complications, the current study covers the whole burden of obesity in the feld of cardiac surgery over short term and long term.

Study Design.
Tis study presents a large single-center, retrospective analysis of all patients who underwent cardiac surgery in the German Heart Center Munich using cardiopulmonary bypass between 2002 and 2017.Data were obtained from an ongoing quality assessment program.All medical reports, including operative protocols, in-hospital, and outpatient notes, were reviewed.

Defnitions
(i) Survival: all-cause mortality was measured.Cox regression analysis was applied for further risk stratifcation for the total observational period and for the assessment of 30-day mortality, logistic regression analyses were used.(ii) Wound-healing disorder (WHD): every sternal wound-healing disorder after cardiac surgery requiring surgical care was counted as WHD, including sternal reconstruction.Te infammatory status of the wound was not additionally classifed.(iii) Re-exploration: re-exploration due to bleeding was defned as a life-threatening, major or minor bleeding according to the BARC criteria, which requires surgical intervention.(iv) Long-term ventilation: long-term ventilation was defned as any necessity for postoperative ventilation >24 h.

Results
14.754 consecutive patients who underwent cardiac surgery using cardiopulmonary bypass over 14 years' period were included to the following analysis.Demographics, intraoperative, and postoperative data are provided in Table 1.1).Similar patterns were found for arterial hypertension and pulmonary obstruction.Notably, 11.0% of the patients with cachexia had a stroke in history compared to the total cohort with 4.1% (p � 0.014), and they had also more malignancies (12.8% versus 6.0%, respectively) but this fnding did not reach signifcance (p � 0.077; Table 1).

2
Journal of Obesity

Journal of Obesity
Patients with cachexia or normal BMI were less likely to undergo CABG (16.5% and 27.3% versus 35.8% of the total cohort, respectively; p < 0.001) but had substantially more valve surgery (49.5% and 38.6% versus 31.7% of the total cohort; p < 0.001).Te CPB time, however, as a surrogate parameter for the complexity of the surgical procedure did not difer between the BMI classes (Table 1 2(a)).Furthermore, no association between BMI classes and 30-day mortality was observed although cachexia had substantially higher odds for mortality but this fnding did not reach signifcance (OR � 2.162; CI 95% [0.941-4.967];p � 0.069).

Wound Healing.
Sternotomy was performed in 13.670 patients (92.7%) and these patients were further analyzed with regards to wound healing.Obesity class I-III had their most impressive impact on the outcome of wound healing.Figure 2(a) demonstrates signifcantly higher percentages of WHD or sternal reconstruction for patients with obesity I-III compared with the total cohort.Multiple logistic regression analyses for the risk of sternal reconstruction (Table 3(a)) and WHD (Table 3 Factors that had a signifcant impact on sternal reconstructions or WHDs were further analyzed, and the forest plots of their odds ratios are given in Figures 2(b) and 2(c), respectively.Notably, the highest odd for sternal reconstruction depicts re-exploration during the postoperative period (OR � 4.535; CI 95% [3.337-6.156];p < 0.001) followed by obesity I-III.Te highest odds for WHD, however, was observed in obesity III (OR � 4.123; CI 95% [2.521-6.743];p < 0.001) just ahead of re-exploration and obesity I-II.

Perioperative Outcome.
Analyzing major adverse events during the short-term, multiple logistic regression analyses were performed.Demographic data as well as comorbidities and surgical parameters were included in the multiple logistic regression analysis.Neither cachexia nor obesity I-III were associated with an increased risk for renal replacement therapy (

Discussion
Obesity is a well-established risk factor for cardiovascular disease, diabetes mellitus, and hypertension [20] and is likely to play an increasingly important role in cardiac surgery in the future.

Survival and Perioperative Outcome.
Little is known about the efect of obesity on long-term outcome after cardiac surgery with a follow-up of more than one year [8].Former research suggested that patients with obesity or at least overweight might have a survival beneft, culminating in the term "obesity paradox" that states that at least mildly obese patients with heart failure might have better clinical outcome than expected [4,7,21,22].Moreover, Zhang et al. suggested a survival beneft of patients with BMI>30 kg/m 2 after cardiac surgery in a subgroup of elderly patients [23].Te efects on survival, however, appeared to be small or almost zero and are controversial [8,9,11,19].Te current analysis did not fnd any survival beneft in obesity patients, neither for the 30-day mortality nor for the long-term outcome.Unlike former research, the current study analyzes a long observational period of 15 years (mean survival of 4.360 ± 39 days, Table 1), with no efects of obesity I-III on survival.
No signifcant associations could be observed with major adverse events during the postoperative period (Tables 3(c)-3(f )).Tese fndings limit the suggestions of former research that reported an increased perioperative morbidity in obesity patients [14,24].Te current study did not fnd strong efects of obesity I-III on renal failure but provides some indication for a possible, small risk in obesity III (OR � 1.623; CI 95% [0.949-2.776];p � 0.077; Table 3

(c)). Te same applies for prolonged ventilation (Table 3(d)).
Furthermore, obesity I-III did not come out as a risk factor for cerebrovascular events (Table 3(f )), which is in line with former research [6].Notably, the necessity for re-exploration or the length of CPB times was not increased in obese patients, which might have been assumed given the technically more difcult operational site.Taken together, the impact of obesity I-III on perioperative morbidity is either very small or as the current study suggests, simply does not exist.

Cachexia.
We found that indeed cachexia was an independent risk factor for all-cause mortality (OR � 2.733; CI 95% [1.874-3.988];p < 0.0001; Table 2(a)) as well as a risk factor for prolonged ventilation in the postoperative period 4 Journal of Obesity As we could rule out malignancies as the main reason for this fnding (Table 1), cachexia might represent an unspecifc marker for the end stage of the underlying cardiac disease.Moreover, the impaired survival of cachectic patients may explain some of the benefcial efects of obesity on survival.If cachectic patients were not analyzed separately in the previous studies, the catastrophic efects of cachexia may have biased the results in favor of obese patients, thus misrepresenting an advantage for obesity.

Te Burden of Obesity.
Te current study suggests that the main disadvantage of obesity in patients undergoing cardiac surgery lies in the impaired outcome of wound healing after sternotomy.Former studies showed that obesity is independently associated with an increased risk of postoperative sternal wound infection [16,25], especially with a BMI over 30 kg/m 2 [26].In order to study a real world situation, we counted any surgical intervention due to wound healing as WHD regardless from their infective state.
Te CDC classifcation focuses on the infective state [27] but

Implications of Obesity in Cardiac Surgery.
As cardiac surgery will be confronted with more obese patients in the foreseeable future, strategies to reduce WHDs are urgently needed.Minimally invasive techniques that do not require conventional sternotomy seem to be the most promising options in cardiac surgery in obese patients: In mitral valve surgery, lateral mini thoracotomy becomes more and more common [28], and catheter techniques [29] or robotic surgery are emerging [30].In aortic valve disease, the ongoing discussion about the indications between transcatheter aortic valve implantations and conventional approaches via sternotomy [31] should be [32] extended [33] and obese [34] patients should be treated.

Conclusions
Tere is no increased risk of short-or long-term mortality after cardiac surgery in overweight patients compared with normal-weight patients.In addition, the major adverse events during the perioperative course are similar to those in normal-weight patients.Te major adverse efect of obesity is the signifcantly increased rate of wound healing disorders and sternal reconstruction.Unexpectedly, patients with cachexia without apparent oncologic disease have a signifcantly increased risk of both the occurrence of perioperative complications and increased short-and long-term mortality.
Furthermore, we analyzed the subgroup of patients that required sternal reconstruction.Taken together, every additional point of BMI worsens the outcome of wound healing (Figure 2(a)), and the risk for sternal reconstructions or WHDs doubled at least in obesity patients (Tables 3(a) and 3(b)).

Table 2 :
(a)Cox regression analysis for all-cause mortality.(b) Logistic regression analysis for 30-day mortality.

Table 3 :
(a) Multiple logistic regression analysis for sternal reconstruction.(b) Multiple logistic regression analysis for wound-healing disorders.(c) Multiple logistic regression analysis for renal replacement therapy.(d) Multiple logistic regression analysis for ventilation >24 h.(e) Multiple logistic regression analysis for re-exploration.(f ) Multiple logistic regression analysis for cerebrovascular events.