Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 May 30.
Published in final edited form as: J Diabetes Complications. 2015 Aug 4;29(8):1177–1182. doi: 10.1016/j.jdiacomp.2015.07.027

The impact of hyperglycemia and obesity on hospitalization costs and clinical outcome in general surgery patients

Lauren Buehler a, Maya Fayfman a, Anastasia-Stefania Alexopoulos a, Liping Zhao b, Farnoosh Farrokhi a, Jeff Weaver c, Dawn Smiley-Byrd a, Francisco J Pasquel a, Priyathama Vellanki a, Guillermo E Umpierrez a,*
PMCID: PMC10227768  NIHMSID: NIHMS1899275  PMID: 26355027

Abstract

Background:

The impact of obesity on clinical outcomes and hospitalization costs in general surgery patients with and without diabetes (DM) is unknown.

Materials and Methods:

We reviewed medical records of 2451 patients who underwent gastrointestinal surgery at two university hospitals. Hyperglycemia was defined as BG ≥140 mg/dl. Overweight was defined by body mass index (BMI) between 25–29.9 kg/m2 and obesity as a BMI ≥30 kg/m2. Hospital cost was calculated using cost-charge ratios from Centers for Medicare and Medicaid Services. Hospital complications included a composite of major cardiovascular events, pneumonia, bacteremia, acute kidney injury (AKI), respiratory failure, and death.

Results:

Hyperglycemia was present in 1575 patients (74.8%). Compared to patients with normoglycemia, those with DM and non-DM with hyperglycemia had higher number of complications (8.9% vs. 35.8% vs. 30.0%, p<0.0001), longer hospital stay (5 days vs. 9 days vs. 9 days, p<0.0001), more readmissions within 30 days (9.3% vs. 18.8% vs. 17.2%, p<0.0001), and higher hospitalization costs ($20,273 vs. $79,545 vs. $72,675, p<0.0001). In contrast, compared to normal-weight subjects, overweight and obesity were not associated with increased hospitalization costs ($58,313 vs. $58,173 vs. $66,633, p=0.74) or risk of complications, except for AKI (11.9% vs. 14.8% vs. 20.5%, p<0.0001). Multivariate analysis revealed that DM (OR=4.4, 95% CI= 2.8,7.0) or perioperative hyperglycemia (OR=4.1, 95% CI=2.7–6.2) were independently associated with increased risk of complications.

Conclusion:

Hyperglycemia but not increasing BMI, in patients with and without diabetes undergoing gastrointestinal surgery was associated with a higher number of complications and hospitalization costs.

Keywords: Type 2 diabetes, Inpatient hyperglycemia, Obesity, Obesity paradox, Hospitalization costs

1. Introduction

Diabetes mellitus (DM) and obesity are leading causes of morbidity and mortality and represent growing burdens on the U.S. health care system. Currently, there are over 29 million Americans living with DM, and over a third of the American population is obese (Ogden, Carroll, Kit, & Flegal, 2014). Among surgical patients, DM and hyperglycemia are associated with perioperative complications, length of hospital stay (LOS), resource utilization, and mortality (Kwon et al., 2013; Noordzij et al., 2007; Mraovic et al., 2010). In addition, several studies and meta-analyses have shown that the development of hyperglycemia, in patients with and without diabetes, is independently associated with complications including pneumonia, bacteremia, urinary tract infection, skin infections, pulmonary embolism, acute kidney injury, and death (Abdelmalak et al., 2014; Frisch & Smiley, 2010; Kwon et al., 2013; Mraovic et al., 2010).

Obesity is widely acknowledged to be a public health threat due to its association with multiple chronic conditions such as DM, hypertension, osteoarthritis, heart disease, some types of cancer (breast, colon, pancreas, and endometrium), and increased cardiovascular disease mortality (Flegal, Graubard, Williamson, & Gail, 2007; Haslam & James, 2005). There is a lack of consensus in the literature regarding the association between BMI and clinical outcome in surgical patients. In a meta-analysis of 18 orthopedic surgery studies (Kluczynski, Bisson, & Marzo, 2014), half of the studies found increased BMI to be associated with worse postoperative outcomes. Similarly, in hospitalized patients with severe trauma, obesity was reported to be associated with prolonged hospital stay and increased risk of complications (Bardou, Barkun, & Martel, 2013; Liu, Chen, Bai, Zheng, & Gao, 2013). In contrast, other studies have shown that obesity may actually be protective against morbidity and mortality in surgical and critically ill patients, a phenomenon which has been dubbed ‘the obesity paradox (Hutagalung et al., 2011; Schwann et al., 2001; Utzolino, Ditzel, Baier, Hopt, & Kaffarnik, 2014; Valentijn et al., 2013; Yamamoto et al., 2013). Despite the lack of a consensus in the literature on the association between obesity and postoperative complications, several studies have shown that surgeons are less likely to operate on obese patients compared to their normal weight counterparts, in particular in patients with diabetes (Reeves, Ascione, Chamberlain, & Angelini, 2003; Wang, Ramanathan, Stewart, Gamble, & White, 2013). Given the inconsistencies and gaps in evidence regarding the association between obesity and surgical outcomes, as well as the lack on information on the impact of diabetes in obese patients undergoing general surgery, we analyzed the effect of hyperglycemia and obesity on clinical outcomes and costs in patients undergoing gastrointestinal surgery.

2. Methods

This study analyzed prospectively gathered data on 2451 patients who underwent intra-abdominal surgery at Emory University Hospitals in Atlanta, Georgia between December 2010 and December 2012. After exclusion of patients who had missing information on BMI, hospital charges or blood glucose measures, a total of 2104 patients were included in the analysis. We included the 10 most commonly performed open and laparoscopic abdominal surgeries at our institution: appendectomy, cholecystectomy, gastric bypass, gastrectomy, colectomy, colostomy revision, hepatectomy, splenectomy, pancreatectomy, and small bowel excision/repairs. Patient selection and type of surgical procedures were identified using Current Procedural Terminology (CPT) codes established by the American Medical Association (AMA) (American Medical Association, 2013). The study was approved by the Emory University Institutional Review Board.

Inpatient hyperglycemia was defined by American Diabetes Association and Endocrine Society guidelines, as any BG value ≥140 mg/dL (7.8 mmol/L)(Moghissi, Korytkowski, DiNardo et al., 2009; Umpierrez et al., 2012) either in the operating room or during the hospital stay, up to seven days postoperatively. For the BMI analysis, BMI was calculated as weight (kg) divided by height squared (m2), and patients were categorized into three groups: normal weight (BMI 18.5–24.9 kg/m2), overweight (BMI 25–29.9 kg/m2), and obese (BMI ≥30 kg/m2). We excluded patients who were underweight (BMI <18.5 kg/m2) as these patients made up <5% of our study population (n=107). The main outcomes for this study included inpatient complications, length of hospital stay, resource utilization, and cost of hospitalization. Prior diagnoses of DM and hospital complications were identified by ICD-9 codes generated during the hospitalization. Complications included a composite of major cardiovascular events, pneumonia, bacteremia, acute kidney injury (AKI), wound infection, and in-hospital mortality. ICD-9 codes were used to determining the presence of complications for all of these except for acute kidney injury, which was defined as an increase in serum creatinine greater than 50% from the patient’s baseline admission creatinine (Bagshaw et al., 2010).

Total hospital cost was calculated from total hospitalization charges using the 2011 and 2012 cost-charge ratios from Centers for Medicare & Medicaid Services. Data identified for extraction included ICD-9, CPT, MS-DRG codes as well as resource utilization data and related hospital charges. These were obtained from the electronic health record and billing/coding departments at Emory University Hospital.

2.1. Statistical analysis

Descriptive statistics were used to examine the distribution of patient demographics and clinical characteristics in the overall study population. Bivariate analyses comparing baseline demographics, complications, and cost by glycemic status and BMI category were performed using χ2 test for categorical variables, ANOVA test for continuous variables with Gaussian distribution, and the Kruskal-Wallis test for continuous nonparametric data. Multiple logistic regression models adjusted for age, gender, race, and type of surgery were used to determine odds ratios and 95% confidence intervals for the relationship between hyperglycemia, BMI, and complications. An interaction term for BMI and glycemic status was included in the model to assess for a potential additive effect between these two variables. Hospital costs were modeled with gamma distribution to estimate adjusted costs by glycemic status or BMI category. Two-sided p<0.05 was considered statistically significant. All analyses were performed using Statistical Analysis Software (SAS) version 9.4.

3. Results

Of the 2104 patients included in the analysis, 453 (21.5%) had a documented diagnosis of diabetes prior to surgery, and 1651 (78.5%) had no known history of diabetes. Of the patients without diabetes, 1122 (68.0%) had ≥1 blood glucose value greater than 140 mg/dL during their hospital stay. The mean BMI of all participants was 28.3 ± 6.6 kg/m2. A total of 716 patients (34.0%) were normal weight, 702 (33.4%) were overweight, and 686 (32.6%) were obese. The mean age of the study population at the time of admission was 55.6 ± 15.0 years, and 1052 (49.5%) of patients were male.

Compared to non-diabetic patients with normoglycemia, patients with DM and non-DM with hyperglycemia were older (48 vs. 60 vs. 57 years, p<0.0001) and had a greater percentage of male subjects (44.0% vs. 57.4% vs. 49.7%, p=0.0002) (Table 1). Patients with DM had a higher mean BMI compared to patients without DM with and without hyperglycemia, respectively (30.0 vs. 28.0 vs. 27.6 kg/m2, p<0.0001). In addition, patients with DM and non-DM patients with hyperglycemia had a longer median length of stay, a higher likelihood of admission to the ICU, and a higher risk for readmission within 30 days after discharge, (all p<0.0001) (Table 1).

Table 1.

Patient demographic and clinical characteristics by glycemic status.

No diabetes
 No diabetes
 Diabetes (n=453) p value
Normoglycemia (n=529) Hyperglycemia (n=1122)
Age, years 48.2 ± 16.0 57.4 ± 14.3 59.8 ± 12.3 <0.0001
Male gender, n (%) 234 (44.2) 558 (49.7) 260 (57.4) 0.0002
Race
 Caucasians, n (%) 332 (68) 766 (73.3) 275 (63.4) 0.0022
 African Americans, n (%) 141 (28.8) 252 (24.1) 147 (33.9)
Hispanic, n (%) 12 (3.1) 16 (2.0) 11 (3.0) 0.3879
BMI, kg/m2 27.6 ± 5.8 28.0 ± 6.6 30.0 ± 7.2 <0.0001
 Normal weight 194 (37) 403 (35.9) 119 (26.3) <0.0001
 Overweight 185 (35) 375 (33.4) 142 (31.3)
 Obese 150 (28) 344 (30.7) 192 (42.4)
ICU, n (%) 26 (4.9) 500 (44.6) 216 (47.7) <0.0001
Open surgery, n (%) 303 (57.3) 903 (80.5) 337 (74.4) <0.0001
Mean admission BG, mmol/L 109.8 ± 15.8 136.7 ± 38.6 162.0 ± 53.4 <0.0001
Mean in-hospital BG, mmol/L 106.3 ± 11.5 128.8 ± 18.9 151.5 ± 28.2 <0.0001
Mild hypoglycemia, n (%) 15 (2.8) 136 (12.1) 102 (22.5) <0.0001
Severe hypoglycemia, n (%) 8 (1.51) 32 (2.9) 14 (3.1) 0.0004
LOS, days 5.0 (3.0–7.0) 9.0 (6.0–15.0) 9.0 (6.0–18.0) <0.0001
Discharge to home, n (%) 515 (97) 978 (87.2) 366 (80.8) <0.0001
Readmission in 30 days, n (%) 49 (9.3) 193 (17.2) 85 (18.8) <0.0001
Complications
 Composite, n (%) 45 (8.5) 327 (29.1) 166 (36.6) <0.0001
 In-hospital Mortality, n (%) 1 (0.2) 34 (3.0) 19 (4.2) 0.0001
 Acute MI, n (%) 1 (0.2) 22 (2.0) 10 (2.2) 0.0121
 Wound infection, n (%) 19 (3.6) 120 (10.7) 53 (11.7) <0.0001
 Pneumonia, n (%) 8 (1.5) 108 (9.6) 50 (11.0) <0.0001
 Sepsis, n (%) 1 (0.2) 5 (0.4) 3 (0.7) 0.5218
 Acute kidney injury, n (%) 17 (3.2) 204 (18.2) 114 (25.2) <0.0001
Total hospitalization Costs, USD 20,273 (18,748, 21,798) 72,675 (65,895, 79,455) 79,545 (67,258, 91,833) <0.0001
Open in a new tab

Composite complications were approximately four times more likely among patients with DM (OR=4.49, 95% CI=2.85, 7.07) and non-DM patients with hyperglycemia (OR=4.08 95% CI=2.69, 6.18) compared to normoglycemia patients (both p<0.0001) (Table 1). Individual complications that were significantly more common among DM and non-DM patients with hyperglycemia include major cardiovascular events, wound infection, pneumonia, AKI and death (all, p<0.05) (Table 1).

Table 2 summarizes the comparison of patient demographic by BMI category (normal weight, overweight, and obese). Obese subjects were younger (54.3 vs. 56.6 vs. 55.8 years, p=0.0122) and were more likely to be female (p<0.0001) than patients in the overweight and normal weight categories. There were no significant differences in length of hospital stay (p=0.43) or likelihood of readmission within 30 days (p=0.39) among the three groups. Likewise, there was no difference in the risk of complications when stratifying by BMI category (all p-values >0.05), except for AKI (p<0.0001). Lastly, mean total cost of hospital stay was higher in the obese category of patients, but this was not statistically significant (p=0.74).

Table 2.

Patient demographic and clinical characteristics by BMI category.

Normal weight (n=716) Overweight (n=702) Obese (n=686) p value
Age, years 55.8 ± 16.3 56.6 ± 14.5 54.3 ± 14.1 0.0122
Male gender, n (%) 345 (48.2) 402 (57.3) 305 (44) <0.0001
Race
 Caucasians, n (%) 476 (72.1) 467 (70.3) 430 (67) 0.25
 African Americans, n (%) 164 (24.9) 180 (27.1) 196 (30.4)
Hispanic, n (%) 12 (2.3) 11 (2.1) 16 (3.1) 0.51
DM/hyperglycemia, n (%)
 Diabetic 119 (16.6) 142 (20.2) 192 (28.0) <0.0001
 Non-diabetic w/BG >=140 403 (56.3) 375 (53.4) 344 (50.2)
 Non-diabetic w/BG <140 194 (27.1) 185 (26.4) 150 (21.9)
ICU, n (%) 233 (32.5) 252 (35.9) 257 (37.5) 0.14
Open surgery, n (%) 547 (76.4) 514 (73.2) 482 (70.3) 0.0342
Mean admission BG, mmol/L 132.8 ± 37.0 138.1 ± 43.1 144.9 ± 49.3 0.001
Mean in-hospital BG, mmol/L 122.7 ± 20.7 128.9 ± 25.3 132.7 ± 28.1 <0.0001
Mild hypoglycemia, n (%) 105 (14.7) 72 (10.3) 76 (11.1) 0.0251
Severe hypoglycemia, n (%) 14 (2.0) 16 (2.3) 16 (2.3) 0.87
LOS, days 8.0 (5.0–14.0) 8.0 (5.0–14.0) 7.0 (5.0–12.0) 0.431
Discharge to home, n (%) 637 (89.0) 626 (89.2) 596 (86.9) 0.33
Readmission in 30 days, n (%) 116 (16.2) 115 (16.4) 96 (14.0) 0.39
Complications
Composite, n (%) 170 (23.7) 173 (24.6) 202 (29.5) 0.033
 In-hospital Mortality, n (%) 21 (2.9) 15 (2.1) 18 (2.6) 0.63
 Acute MI, n (%) 14 (2.0) 11 (1.6) 8 (1.2) 0.49
 Wound infection, n (%) 63 (8.8) 62 (8.8) 67 (9.8) 0.78
 Pneumonia, n (%) 66 (9.2) 49 (7.0) 51 (7.4) 0.26
 Sepsis, n (%) 4 (0.6) 3 (0.4) 2 (0.3) 0.75
 Acute kidney injury, n (%) 85 (11.9) 104 (14.8) 140 (20.5) <0.0001
Total hospitalization Costs, USD 58,313 (50,906, 65,720) 58,173 (52,226, 64,120) 66,633 (56,489, 76,776) 0.74
Open in a new tab

A logistic regression model adjusted for age, sex, race and type of surgery was used to determine odds ratios comparing non-DM patients with normoglycemia to non-DM patients with hyperglycemia and DM patients within each of the three BMI categories. Overall, the presence of either DM (OR=4.5, 95% CI=2.9–7.1) or perioperative hyperglycemia (OR=4.1, 95% CI=2.7–6.2) was associated with over a four-fold increase in odds of complications when compared with normoglycemic non-DM patients. When stratifying by BMI category, a prior diagnosis of DM was associated with a five-fold increase in odds of complications in the normal weight group (OR=5.2, 95% CI=2.6–10.1), a four-fold increase in the overweight category (OR=4.3, 95% CI=2.2–8.2), and a six-fold increase in the obese category (OR=6.0, 95% CI=3.0–12.1) compared to non-DM patients with normoglycemia (Appendix Table 1).

Non-DM patients with hyperglycemia had higher rates of perioperative complications compared to non-DM subjects with normoglycemia when stratified by BMI group [(normal weight: OR=3.1, 95% (CI=1.7–5.7), overweight group: OR=4.2, 95% (CI=2.3–7.6), and obese group: OR=5.8, 95% (CI=3.0–11.2)], (Appendix Table 1). In contrast, a logistic model controlling for age, sex, race and type of surgery found no significant increase in the odds of postoperative complications when comparing the overweight and normal weight groups within each glycemic category. However, there was a slightly increased odds of complications when comparing obesity to normal weight in the group who were non-DM with hyperglycemia (OR 1.7, 95% CI=1.2–2.3) (Appendix Table 2).

Adjusted total cost of hospital stay was significantly lower in those patients who had no prior history of DM and who did not have hyperglycemia during the perioperative period. The mean total cost of hospitalization for non-DM patients with normoglycemia was $20,273 compared to $72,675 for non-DM patients with hyperglycemia and $79,545 for patients with a prior diagnosis of DM (p<0.0001) (Fig. 1). In contrast, there was no significant difference in the adjusted total cost of hospitalization among the three BMI categories (p=0.74). The mean total cost of hospital stay was $58,313 for normal weight individuals, $58,173 in the overweight category, and $66,633 in the obese group (p=0.41) (Fig. 2).

Fig. 1.

Fig. 1.

Open in a new tab

Hospital complications and hospitalization costs by glycemic status in patients undergoing abdominal surgery. Composite of hospital complications (1A) and hospitalization costs (1B) in non-diabetic patients with normoglycemia, non-diabetic patients with hyperglycemia and diabetes. Hyperglycemia is defined as a blood glucose > 140 mg/dl. Composite of complication included mortality, wound infection, pneumonia, bacteremia, respiratory failure, acute kidney injury and major cardiovascular events. DM=Diabetes. * p<0.0001.

Fig. 2.

Fig. 2.

Open in a new tab

Hospital complications and hospitalization costs by body mass index (BMI) status in patients undergoing intra-abdominal surgery. Composite of hospital complications (2A) and hospitalization costs (2B)in normal weight (18.5–25 kg/m2), overweight (25–30 kg/m2) and obese (>30 kg/m2) patients. Composite of complication included mortality, wound infection, pneumonia, bacteremia, respiratory failure, acute kidney injury and major cardiovascular events. DM=Diabetes.

4. Discussion

This study aimed to assess whether DM, perioperative hyperglycemia, and obesity are independent predictors of adverse postoperative outcomes and increased hospital cost in patients undergoing general surgery. We found that diabetes and perioperative hyperglycemia were associated with a higher number of postoperative complications, higher in-hospital mortality, longer length of stay, increased likelihood of readmission within 30 days, and higher total cost of hospitalization. In contrast, the presence of overweight and obesity was not associated with a significant increase in complications, hospitalization cost, or resource utilization compared to normal weight subjects.

The results of our study confirm the strong association between DM and perioperative hyperglycemia and poor surgical outcome. Several retrospective and randomized controlled trials in cardio-surgical, ICU (Abdelmalak et al., 2014; American Medical Association, 2013; Bagshaw et al., 2010; Baker et al., 2006; Bardou et al., 2013; Flegal et al., 2007; Frisch & Smiley, 2010; Haslam & James, 2005; Hutagalung et al., 2011; Kluczynski et al., 2014; Liu et al., 2013; Moghissi et al., 2009; Schwann et al., 2001; Umpierrez et al., 2012; Utzolino et al., 2014; Valentijn et al., 2013; Yamamoto et al., 2013), and general surgery populations have reported that hyperglycemia and diabetes are associated with increased length of stay, hospital complications and mortality, and that improvement in glycemic control reduces hospital complications (Baker et al., 2006; Clement et al., 2004; McAlister et al., 2005; Pomposelli et al., 1998; Umpierrez et al., 2002), as well a significant increase in the risk for mortality within one year of surgery (Abdelmalak et al., 2014).

A recent report by the American Diabetes Association (ADA) estimated that the national cost of diabetes care in 2012 was $245 billion, of which $176 billion (72%) represented direct medical costs related to diabetes (American Diabetes A, 2013). Inpatient care represents more than 40% ($76 billion) of the total medical cost of diabetes care (American Diabetes A, 2013). The increased cost of diabetes care relates to the high volume of patients with diabetes (American Diabetes A, 2013), but the presence of perioperative complications was recently shown to play a substantial role in the inpatient care of diabetes (Frisch et al., 2010; Furnary, Zerr, Grunkemeier, & Starr, 1999). Eappen et al. reported that the occurrence of postsurgical hospital complications was associated with a $39,017 higher contribution margin per patient with private insurance and a $1749 higher contribution margin per patient with Medicare compared with absence of postsurgical complications (Eappen et al., 2013). In agreement with these studies, we found that the rate of the composite of complications was approximately four times more likely among patients with diabetes and non-DM patients with perioperative hyperglycemia compared to normoglycemic non-DM patients. The hospitalization costs was three-fold higher (greater than $40,000 per case) in non-DM patients with hyperglycemia and those with prior history of DM, compared to non-DM subjects without hyperglycemia.

In contrast to our findings regarding perioperative hyperglycemia, we found that overweight and obesity (defined by BMI) had no effect on postoperative outcome, need for ICU admission, and cost of hospitalization. The only complication that was significantly different by BMI category was AKI, which was more common among obese patients. The currently available literature regarding the impact of obesity on cost and complication rate is highly inconsistent, and there have been few studies using general surgery patients. The complication that has been most consistently linked to obesity in a variety of surgical types is surgical site infection; however, our study failed to show an effect of BMI on this complication (Dindo, Muller, Weber, & Clavien, 2003; Kim et al., 2003; Kuduvalli, Grayson, Oo, Fabri, & Rashid, 2002; Schwann et al., 2001; Yuan & Chen, 2013). This is likely because the majority of evidence on obesity and surgical outcomes focuses on patients undergoing cardiac surgery (Cemerlic-Adjic et al., 2014; Kim et al., 2003; Kuduvalli et al., 2002; Pan, Hindler, Lee, Vaughn, & Collard, 2006; Wang et al., 2013), while our study was limited to patients undergoing abdominal surgery. A large study on cardiac surgical patients by Wagner et al. showed a U-shaped curve for risk of complications and mortality by BMI for patients undergoing CABG, suggesting that both underweight and obese patients are at greater risk for morbidity and mortality compared to normal weight subjects (Wagner et al., 2007). Similar findings were reported by Zhao et al., who found that all-cause mortality among patients with DM2 followed a U-shaped curve with greatest mortality at the very low and high extremes of BMI (Zhao et al., 2014).

Similar to the results in our study, other studies from a variety of surgical populations have found no association between BMI and adverse postoperative events, including a large cohort study of nearly 150,000 general surgery patients that showed BMI was not a significant predictor of postoperative complications or 30-day mortality (Cemerlic-Adjic et al., 2014) While our study did not show this, some data suggest that obesity may actually be protective against complications and mortality, a phenomenon which has been called the “obesity paradox.” Hutagalung et al., 2011; Reeves et al., 2003; Schwann et al., 2001; Utzolino et al., 2014; Valentijn et al., 2013; Yamamoto et al., 2013 For example, in a large cohort study on over 12,000 surgical ICU patients, obesity was associated with lower risk of 60-day in-hospital mortality (Hutagalung et al., 2011). However, one of the largest meta-analyses published on the subject, which included over 2.88 million patients, showed no evidence of an obesity paradox, but rather an increase in complications in high grade obesity (Flegal, Kit, Orpana, & Graubard, 2013).

Although BMI was not a significant predictor of most postoperative complications in the logistic model including all study participants, it is important to note that obesity was associated with a higher rate of complications in non-DM patients with hyperglycemia. In this group, the presence of obesity compared to normal weight was associated with a 1.7 fold increased odds of complications after surgery (95% CI=1.2–2.3). These findings suggest a potential additive effect of obesity and perioperative stress hyperglycemia in non-DM patients.

Among the complications addressed, AKI was the only complication that was significantly more common among obese patients compared to their normal weight and overweight counterparts. This finding corroborates those of several other studies showing that higher BMI is associated with increased risk for AKI in the postoperative setting (Kelz et al., 2013; Schmid et al., 2014). A possible explanation may be that obesity may predispose patients to AKI postoperatively because volume assessment and resuscitation is more difficult in the obese patient, which places patients at risk for pre-renal AKI (Kelz et al., 2013).

There are several limitations to this study. While we adjusted for several covariates in our analysis such as age, sex, type of surgery (laparoscopic vs. open), we were not able to adjust for other confounding factors such as coexisting comorbidities, medication use, and severity of illness prior to surgery. Additionally, there may be unknown confounding variables that may have affected our results. In addition, we limited our investigation to patients undergoing intra-abdominal surgery, thus our results need to be confirmed in other surgical populations.

In summary, our study provides novel and important clinical implications for the management of general surgery patients. Our results indicate that diabetes and perioperative hyperglycemia were associated with a higher number of postoperative complications, higher in-hospital mortality, and higher total cost of hospitalization. In contrast, obesity has limited impact on key postoperative outcomes and cost, which should dissuade surgeons from withholding surgery from obese patients on this basis. Prospective randomized clinical studies are needed to determine best treatment strategies for the management of obese subjects with hyperglycemia and diabetes undergoing general surgery.

Acknowledgements

The present study was supported by an unrestricted grant from the Jacobs Family Research Fund (to Emory University and GEU). Dr. Umpierrez is supported in part by research grants from the American Diabetes Association (1-14-LLY-36), and PHS Grant UL1 RR025008 from the Clinical and Translational Science Award program, National Institutes of Health, National Center for Research Resources. L.B., M.F., A.S.A., L.Z., F.F., J.W., D.S-B. F.J.P., P.V. have no COI to declare.

Appendix A

Appendix Table 1.

Effect of diabetes and hyperglycemia on complications by BMI group.

Odds ratio (95% CI)
Normal
weight		 Overweight Obese
DM vs. Non-DM with normoglycemia 5.2 (2.6–10.1) 4.3 (2.2–8.2) 6.0 (3.0–12.1)
Non-DM with hyperglycemia vs. Non-DM with normoglycemia 3.1 (1.7–5.7) 4.2 (2.3–7.6) 5.8 (3.0–11.2)
Open in a new tab

Appendix Table 2.

Effect of BMI and composite on complications by glycemic status.

Odds Ratio (95% CI)
No-DM
normoglycemia		 No-DM
hyperglycemia		 DM
Overweight vs. normal weight 0.9 (0.4–1.9) 1.2 (0.9–1.7) 0.8 (0.4–1.3)
Obese vs. normal weight 0.7 (0.3–1.7) 1.7 (1.2–2.3) 1.0 (0.6–1.5)
Open in a new tab

Footnotes

Conflicts of Interest Statements:

• Guillermo Umpierrez is supported in part by research grants from the American Diabetes Association (1-14-LLY-36), and PHS Grant UL1 RR025008 from the Clinical and Translational Science Award program, National Institutes of Health, National Center for Research Resources.

• Lauren Buehler – Conflicts of interest: none

• Maya Fayfman – Conflicts of interest: none

• Anastasia-Stefania Alexopoulos – Conflicts of interest: none

• Liping Zhao – Conflicts of interest: none

• Farnoosh Farrokhi – Conflicts of interest: none

• Jeff Weaver – Conflicts of interest: none

• Dawn Smiley-Byrd – Conflicts of interest: none

• Francisco J. Pasquel – Conflicts of interest: none

• Priyathama Vellanki – Conflicts of interest: none

References

  1. Abdelmalak BB, Knittel J, Abdelmalak JB, Dalton JE, Christiansen E, Foss J, et al. (2014). Preoperative blood glucose concentrations and postoperative outcomes after elective non-cardiac surgery: an observational study. British Journal of Anaesthesia, 112(1), 79–88. [DOI] [PubMed] [Google Scholar]
  2. American Diabetes A (2013). Economic costs of diabetes in the U.S. in 2012. Diabetes Care, 36(4), 1033–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. American Medical Association (2013). Current Procedural Terminology. (2013). [Google Scholar]
  4. Bagshaw SM, Bellomo R, Devarajan P, Johnson C, Karvellas CJ, Kutsiogiannis DJ, et al. (2010). Review article: Acute kidney injury in critical illness. Canadian Journal of Anaesthesia = Journal Canadien d'Anesthésie, 57(11), 985–998. [DOI] [PubMed] [Google Scholar]
  5. Baker EH, Janaway CH, Philips BJ, Brennan AL, Baines DL, Wood DM, et al. (2006). Hyperglycaemia is associated with poor outcomes in patients admitted to hospital with acute exacerbations of chronic obstructive pulmonary disease. Thorax, 61(4), 284–289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bardou M, Barkun AN, & Martel M (2013). Obesity and colorectal cancer. Gut, 62(6), 933–947. [DOI] [PubMed] [Google Scholar]
  7. Cemerlic-Adjic N, Pavlovic K, Jevtic M, Velicki R, Kostovski S, & Velicki L (2014). The impact of obesity on early mortality after coronary artery bypass grafting. Vojnosanitetski Pregled. Military-Medical and Pharmaceutical Review, 71 (1), 27–32. [DOI] [PubMed] [Google Scholar]
  8. Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG, et al. (2004). Management of diabetes and hyperglycemia in hospitals. Diabetes Care, 27(2), 553–591. [DOI] [PubMed] [Google Scholar]
  9. Dindo D, Muller MK, Weber M, & Clavien PA (2003). Obesity in general elective surgery. Lancet, 361(9374), 2032–2035. [DOI] [PubMed] [Google Scholar]
  10. Eappen S, Lane BH, Rosenberg B, Lipsitz SA, Sadoff D, Matheson D, et al. (2013). Relationship between occurrence of surgical complications and hospital finances. JAMA : The Journal of the American Medical Association, 309(15), 1599–1606. [DOI] [PubMed] [Google Scholar]
  11. Flegal KM, Graubard BI, Williamson DF, & Gail MH (2007). Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA, 298(17), 2028–2037. [DOI] [PubMed] [Google Scholar]
  12. Flegal KM, Kit BK, Orpana H, & Graubard BI (2013). Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis. JAMA, 309(1), 71–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Frisch A, Chandra P, Smiley D, Peng L, Rizzo M, Gatcliffe C, et al. (2010). Prevalence and clinical outcome of hyperglycemia in the perioperative period in noncardiac surgery. Diabetes Care, 33(8), 1783–1788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Frisch ACP, & Smiley D (2010). Prevalence and clinical outcome of hyperglycemia in the perioperative period in noncardiac surgery. Diabetes Care, 33(8), 1783–1788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Furnary AP, Zerr KJ, Grunkemeier GL, & Starr A (1999). Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures. Annals of Thoracic Surgery, 67(2), 352–360 (discussion 360–352). [DOI] [PubMed] [Google Scholar]
  16. Haslam DW, & James WP (2005). Obesity. Lancet, 366(9492), 1197–1209. [DOI] [PubMed] [Google Scholar]
  17. Hutagalung R, Marques J, Kobylka K, Zeidan M, Kabisch B, Brunkhorst F, et al. (2011). The obesity paradox in surgical intensive care unit patients. Intensive Care Medicine, 37(11), 1793–1799. [DOI] [PubMed] [Google Scholar]
  18. Kelz RR, Reinke CE, Zubizarreta JR, Wang M, Saynisch P, Even-Shoshan O, et al. (2013). Acute kidney injury, renal function, and the elderly obese surgical patient: a matched case–control study. Annals of Surgery, 258(2), 359–363. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kim J, Hammar N, Jakobsson K, Luepker RV, McGovern PG, & Ivert T (2003). Obesity and the risk of early and late mortality after coronary artery bypass graft surgery. American Heart Journal, 146(3), 555–560. [DOI] [PubMed] [Google Scholar]
  20. Kluczynski MA, Bisson LJ, & Marzo JM (2014). Does body mass index affect outcomes of ambulatory knee and shoulder surgery? Arthroscopy: the Journal of Arthroscopic & Related Surgery: Official Publication of the Arthroscopy Association of North America and the International Arthroscopy Association, 30(7), 856–865. [DOI] [PubMed] [Google Scholar]
  21. Kuduvalli M, Grayson AD, Oo AY, Fabri BM, & Rashid A (2002). Risk of morbidity and in-hospital mortality in obese patients undergoing coronary artery bypass surgery. European Journal of Cardio-Thoracic Surgery: Official Journal of the European Association for Cardio-Thoracic Surgery, 22(5), 787–793. [DOI] [PubMed] [Google Scholar]
  22. Kwon S, Thompson R, Dellinger P, Yanez D, Farrohki E, & Flum D (2013). Importance of Perioperative Glycemic Control in General Surgery. Annals of Surgery, 257(1), 8–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Liu T, Chen JJ, Bai XJ, Zheng GS, & Gao W (2013). The effect of obesity on outcomes in trauma patients: a meta-analysis. Injury, 44(9), 1145–1152. [DOI] [PubMed] [Google Scholar]
  24. McAlister FA, Majumdar SR, Blitz S, Rowe BH, Romney J, & Marrie TJ (2005). The relation between hyperglycemia and outcomes in 2,471 patients admitted to the hospital with community-acquired pneumonia. Diabetes Care, 28(4), 810–815. [DOI] [PubMed] [Google Scholar]
  25. Moghissi ES, Korytkowski MT, DiNardo M, Einhorn D, Hellman R, Hirsch IB, et al. (2009). American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Diabetes Care, 32(6), 1119–1131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mraovic B, Hipszer BR, Epstein RH, Pequignot EC, Parvizi J, & Joseph JI (2010). Preadmission hyperglycemia is an independent risk factor for in-hospital symptomatic pulmonary embolism after major orthopedic surgery. Journal of Arthroplasty, 25(1), 64–70. [DOI] [PubMed] [Google Scholar]
  27. Noordzij PG, Boersma E, Schreiner F, Kertai MD, Feringa HH, Dunkelgrun M, et al. (2007). Increased preoperative glucose levels are associated with perioperative mortality in patients undergoing noncardiac, nonvascular surgery. European Journal of Endocrinology, 156(1), 137–142. [DOI] [PubMed] [Google Scholar]
  28. Ogden CL, Carroll MD, Kit BK, & Flegal KM (2014). Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 311(8), 806–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pan W, Hindler K, Lee VV, Vaughn WK, & Collard CD (2006). Obesity in diabetic patients undergoing coronary artery bypass graft surgery is associated with increased postoperative morbidity. Anesthesiology, 104(3), 441–447. [DOI] [PubMed] [Google Scholar]
  30. Pomposelli JJ, Baxter JK III, Babineau TJ, Pomfret EA, Driscoll DF, Forse RA, et al. (1998). Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. JPEN Journal of Parenteral and Enteral Nutrition, 22(2), 77–81. [DOI] [PubMed] [Google Scholar]
  31. Reeves BC, Ascione R, Chamberlain MH, & Angelini GD (2003). Effect of body mass index on early outcomes in patients undergoing coronary artery bypass surgery. Journal of the American College of Cardiology, 42(4), 668–676. [DOI] [PubMed] [Google Scholar]
  32. Schmid M, Abd-El-Barr AE, Gandaglia G, Sood A, Olugbade K Jr., Ruhotina N, et al. (2014). Predictors of 30-day acute kidney injury following radical and partial nephrectomy for renal cell carcinoma. Urologic Oncology, 32(8), 1259–1266. [DOI] [PubMed] [Google Scholar]
  33. Schwann TA, Habib RH, Zacharias A, Parenteau GL, Riordan CJ, Durham SJ, et al. (2001). Effects of body size on operative, intermediate, and long-term outcomes after coronary artery bypass operation. The Annals of Thoracic Surgery, 71(2), 521–530 (discussion 530–521). [DOI] [PubMed] [Google Scholar]
  34. Umpierrez GE, Hellman R, Korytkowski MT, Kosiborod M, Maynard GA, Montori VM, et al. (2012). Management of hyperglycemia in hospitalized patients in non-critical care setting: an endocrine society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism, 97(1), 16–38. [DOI] [PubMed] [Google Scholar]
  35. Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, & Kitabchi AE (2002). Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. Journal of Clinical Endocrinology and Metabolism, 87(3), 978–982. [DOI] [PubMed] [Google Scholar]
  36. Utzolino S, Ditzel CM, Baier PK, Hopt UT, & Kaffarnik MF (2014). The obesity paradox in surgical intensive care patients with peritonitis. Journal of Critical Care, 29(5) (887.e881–885). [DOI] [PubMed] [Google Scholar]
  37. Valentijn TM, Galal W, Hoeks SE, van Gestel YR, Verhagen HJ, & Stolker RJ (2013). Impact of obesity on postoperative and long-term outcomes in a general surgery population: a retrospective cohort study. World Journal of Surgery, 37(11), 2561–2568. [DOI] [PubMed] [Google Scholar]
  38. Wagner BD, Grunwald GK, Rumsfeld JS, Hill JO, Ho PM, Wyatt HR, et al. (2007). Relationship of body mass index with outcomes after coronary artery bypass graft surgery. The Annals of Thoracic Surgery, 84(1), 10–16. [DOI] [PubMed] [Google Scholar]
  39. Wang TK, Ramanathan T, Stewart R, Gamble G, & White H (2013). Lack of relationship between obesity and mortality or morbidity after coronary artery bypass grafting. The New Zealand Medical Journal, 126(1386), 56–65. [PubMed] [Google Scholar]
  40. Yamamoto M, Mouillet G, Oguri A, Gilard M, Laskar M, Eltchaninoff H, et al. (2013). Effect of body mass index on 30- and 365-day complication and survival rates of transcatheter aortic valve implantation (from the FRench Aortic National CoreValve and Edwards 2 [FRANCE 2] registry). The American Journal of Cardiology, 112(12), 1932–1937. [DOI] [PubMed] [Google Scholar]
  41. Yuan K, & Chen HL (2013). Obesity and surgical site infections risk in orthopedics: a meta-analysis. International Journal of Surgery (London, England), 11(5), 383–388. [DOI] [PubMed] [Google Scholar]
  42. Zhao W, Katzmarzyk PT, Horswell R, Wang Y, Li W, Johnson J, et al. (2014). Body mass index and the risk of all-cause mortality among patients with type 2 diabetes mellitus. Circulation, 130(24), 2143–2151. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES