You are here

Article Text

Article menu
Medical management
Original research
Association between obesity and 1-year mortality in septic patients: a retrospective cohort study
  1. Li Zhang1,
  2. Lingna Fang2,
  3. Hang Lihua1,
  4. Chong Li3
  1. 1Department of Anesthesiology, The First People's Hospital of Kunshan Affiliated with Jiangsu University, Kunshan, Jiangsu, People's Republic of China
  2. 2Department of Endocrinology, The First People’s Hospital of Kunshan Affiliated with Jiangsu University, Kunshan, China
  3. 3Department of Osteoporosis, The First People's Hospital of Kunshan Affiliated with Jiangsu University, Kunshan, Jiangsu, People's Republic of China
  1. Correspondence to Professor Chong Li; sdzl7182{at}163.com

Abstract

Objective Sepsis is a major contributor of intensive care units (ICUs) patient mortality. Prior investigations claimed that obesity enhances overall survival (OS) of septic patients. However, the reported results were inconsistent. This study examined the association between obesity and the 1-year mortality of septic patients.

Design A retrospective cohort study.

Setting The Medical Information Mart for Intensive Care III database.

Participants 3145 septic patients were separated into three distinct cohorts, based on their WHO body mass index (BMI) status.

Outcomes Our primary endpoint was the 1-year mortality from the date of ICU hospitalization.

Result 1334 (42.4%) died within 1 year. The 1-year mortality rate was low in obese patients (38.8%), compared with normal (46.9%) and overweight (42.1%) patients. Crude assessment revealed that obese patients experienced reduced 1-year mortality, relative to normal weight patients (HR 0.79, 95% CI 0.69 to 0.9, p<0.001). However, once adjusted for baseline variables and comorbidities, no correlation was found between obesity and the 1-year mortality (HR 0.93, 95% CI 0.81 to 1.06, p=0.28) of septic patients. There was an association among diabetic (HR 0.72, 95% CI 0.56 to 0.93, p=0.012) and hypertensive (HR 0.73, 95% CI 0.58 to 0.92, p=0.008) patients, and among males (HR 0.71, 95% CI 0.59 to 0.86, p<0.001), with obese individuals experiencing the lowest mortality rate. Given these evidences, the interactions between BMI and mortality in diabetic (p=0.031) and hypertensive (p=0.035) patients were significant.

Conclusion In our study, obese diabetic and hypertensive patients associated to less sepsis-related mortality risk, compared with normal weight patients. Further researches were need to validated.

  • Adult intensive & critical care
  • ANAESTHETICS
  • ALTITUDE MEDICINE

Data availability statement

Data are available on reasonable request. Data for this study were obtained from MIMIC III, which was approved by Massachusetts Institute of Technology (No. 0403000206) and Beth Israel Deaconess Medical Center (2001-P-001699/14).

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjopen-2022-066526

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    STRENGTHS AND LIMITATION OF THIS STUDY

    • Large septic patients in Medical Information Mart for Intensive Care III.

    • Univariable analysis shown demographic, comorbidities, vital signs and lab tests associated with mortality.

    • Cox regression was used to calculate the adjusted HR of obesity and 1-year mortality association.

    • The research was retrospective and was based out of data from a single centre, there may be some unintentional biases that influenced the results.

    Introduction

    Sepsis is a widespread and life-threatening condition.1 However, the heterogeneity associated with sepsis and septic shock prevents the collection of reproducible mortality-related data, which generally ranges between 15% and 56%.1–6 Hence, a closer inspection of patient characteristics that enable strong survival outcomes in septic patients may allow establishment of new and effective approaches of managing sepsis.

    Over 25% of adult intensive care units (ICUs) patients in the USA are overweight or obese.7 8 Emerging evidences suggested the presence of an obesity paradox, whereby obesity markedly reduces mortality in terminally ill patients9 10 like those with persistent cardiac disease, renal insufficiency and sepsis.6 11–13 The underlying mechanism behind this is yet undetermined. However, some speculate involvement of enhanced metabolic reserve,14 15 stimulation of the renin–angiotensin axis16 and release of adipose tissue-based immunomodulatory mediators like leptin and soluble-tumour necrosis factor receptor-2.17

    Currently, there is limited information on the link between sepsis and obesity. More importantly, the reported studies are limited by small sample size, and failure to compensate for possible confounders like severity of illness, comorbidities, infection location6 18 or recent weight loss. Notably, any weight measurement following fluid resuscitation can produce false estimate elevations. Similarly, late-stage weight measurements, such as those obtained later during hospitalisation, can also produce false lows, particularly, if the patient experienced weight loss owing to an extended hospital stay or sepsis catabolism. To prevent possible fluid resuscitation-mediated weight gain or septic catabolism-mediated weight loss, we measured the initial body mass index (BMI) data using dry body weight.

    Given the aforementioned challenges, the true relationship between obesity and overall survival (OS) following sepsis remains unknown. This warrants additional research involving large sample population and extensive analyses. To obtain better septic patient prognosis, it is crucial to uncover any potential link between obesity and septic patient OS. In case obesity truly protects against sepsis-related death, then an extensive analysis of the underlying signalling is necessary to identify sepsis aetiology and develop efficacious therapies. Alternately, in case obesity further complicates sepsis-related conditions, including death, then researchers may need to explore novel approaches to treat this subgroup of septic patients. Therefore, here, we rigorously analysed a large clinical data repository to identify the relationship, if any, between BMI and septic patient OS, after adjusting for possible confounders. Hypertensive19 and insulin resistant20 patients may also act as potential protectors, improving survival after the onset of sepsis. When analysing a person’s risk for obesity, it is important to take into account the sex and gender differences in adipose tissue.21 Therefore, we performed a subgroup analysis based on gender, diabetes and hypertension.

    Methods

    Study approval and design

    This investigation followed the guidelines of the STrengthening the Reporting of OBservational studies in Epidemiology statement. The research design was a retrospective, longitudinal, single-centre, retrospective examination of adult patients diagnosed with sepsis.

    Database

    This research received approval from the Massachusetts Institute of Technology and Beth Israel Deaconess Medical Centre (BIDMC) review boards. In addition, the informed consent requirement was waived due to the retrospective nature of the investigation. Medical Information Mart for Intensive Care III (MIMIC-III) provided physiological data of 38 605 ICU patients, who were admitted to the ICUs of BIDMC, a tertiary care university hospital in Boston, Massachusetts, USA between the years 2002 and 2011.

    Patient and public involvement

    Neither patients nor the public were directly involved in this.

    The sample cohort

    The initial patient selection identified all hospitalised patients with a sepsis diagnosis within the aforementioned time period. Among these patients, the following were chosen for analysis: (1) age >18 years (2) admitted to the medical ICUs and SICU (3) first ICU admission (4) available BMI information. The following patients were eliminated from analysis: (1) those with missing height or weight information and (2) BMI <18.5 kg/m2.

    Following a review of 38 605 MIMIC-III adult admissions, we discovered sepsis in 17 420 patients using the Angus methodology.22 We finally select 3145 patients for analysis. Overall, our study population (figure 1) consisted of 1000 normal weight (BMI 18.5 to <25 kg/m2), 976 overweight (BMI 25 to <29.9 kg/m2) and 1169 obese (BMI ≥30 kg/m2) patients.

    Figure 1
    Figure 1

    Flow chart of patient selection. Study cohort. Illustration of exclusion and inclusion criteria as utilised to select the final cohort of 3145 patients. BMI, body mass index; ICU, intensive care unit; CCU, cardiac care unit; CSRU, cardiac surgery recovery room; SICU, surgical intensive care unit; MIMIC, Medical Information Mart for Intensive Care III.

    Data retrieval

    The structure query language (SQL) in PostgreSQL (V.9.6; PostgreSQL Global Development Group) was used to retrieve data. The following parameters were recorded and analysed: baseline profiles (eg, age, sex), vital signs, severity of organ failure 24 hours post-ICU admission (eg, maximum heart rate, minimum arterial pressure, minimum temperature, Sequential Organ Failure Assessment (SOFA), Simplified Acute Physiology Score (SAPS)), intervention-related data (eg, sedative, vasopressor, mechanical ventilation) and laboratory results (eg, haemoglobin, pH, partial pressure of oxygen ­(PO2)). BMI was computed as weight (in kilograms)/height (in metres).2

    Outcomes

    Our primary endpoint was the 1-year mortality from the date of ICU hospitalisation. The US Social Security Death Index provided information regarding patient deaths following discharge.

    Statistical analysis

    All study variables were assessed using univariate analysis. Continuous data, assessed via pairwise Student’s t-tests and one-way ANOVA, are provided as means±SD and medians (IQR). Categorical data, examined using Pearson’s χ2 or Fisher’s exact test, where necessary, are provided as numbers and percentages. Variables that were significant (p values <0.1) in univariate analysis or were deemed clinically relevant were entered into multivariate analysis. We explored the multicollinearity for independent data using the variance inflation factor, with an upper limit of 2 (online supplemental table S1). Missing data are processed using multiple interpolations. Data missing values are given in online supplemental table S2. Lastly, a subgroup analysis was carried out to identify any potential differences among individual subgroups and interaction tests were also performed.

    All analyses were performed using R Statistical Software (http://www.R-project.org, The R Foundation). A two-sided p<0.05 was set as the significance threshold.

    Covariates

    On admission, all demographic and admission data, including age, gender, weight, height and disease severity, were recorded using the SAPS, SOFA and Elixhauser comorbidity scores.23–25

    Comorbidities

    We identified the following comorbidities using the documented international classification of disease-9 (ICD-9) codes: diabetes, hypertension, chronic obstructive pulmonary disease (COPD), persistent renal disease, liver disease, coronary artery disease (CAD), congestive heart failure (CHF), atrial fibrillation (AFIB), stroke and malignant tumour.

    Results

    Baseline characteristics of sepsis cases

    Table 1 lists the patient baseline profiles of 3145 septic adults, based on their matched BMI statuses. The mean age was 65.3±16.6 years and 53.2% were male. Obese patients exhibited younger age (62.9±15.3 vs 67.5±17.3) and there were more obese females (51.8% vs 45.2%) than normal weight patients. Among obese patients, common comorbidities such as diabetes, hypertension, CHF, renal disease and liver disease were markedly elevated, compared with normal weight patients. Similarly, the SOFA scores within 24 hours of ICU admission were also considerably elevated in obese (6.0 (4.0, 9.0) vs 5.0 (3.0, 8.0)) vs normal weight patients.

    View this table:
    Table 1

    Baseline of the cohort

    Outcome

    Table 2 summarises the patient prognoses of all study participants. In total, 772 (24.5%) and 1334 (42.4%) patients died within 28 days and a year after ICU admission, respectively. Moreover, 272 normal weight (27.2%), 244 overweight (25%), and 256 obese (21.9%) patients died within 28 days after ICU admission. In addition, 469 normal weight (46.9%), 411 overweight (42.1%) and 454 obese (38.8%) patients died within 1 year after admission. Lastly, the median ICU stay duration of obese patients was longer (5.1 days (IQR 2.2–12.0)) than normal weight patients.

    View this table:
    Table 2

    One-year mortality in different BMI patients with sepsis

    Obesity and the 1-year mortality rate

    Figure 2 illustrates the univariate analysis of the BMI-stratified 1-year mortality rate of patients with sepsis. Our analyses revealed a marked relationship between patient variables (ie, age, BMI, diabetes, hypertension, CAD, CHF, AFIB, malignant cancer, stroke, chronic obstructive pulmonary disease, renal disease, liver disease, vital signs within 24 hours of ICU admission, SOFA and SAPS) and 1-year mortality. We adjusted for potential confounding factors in subsequent analyses.

    Figure 2
    Figure 2

    Univariable Cox analysis figure baseline of the cohort. Forest plot show the hazard ratio of univariate analysis.

    Considering BMI as a continuous variable, we observed a marked correlated between BMI and 1-year mortality (HR 0.99, 95% CI 0.98 to 0.99, p<0.001). Following additional age and sex adjustment, the relationship altered (HR 0.99, 95% CI 0.99 to 1, p=0.148).

    Based on crude analysis, obese patients exhibited reduced 1-year mortality, relative to normal weight patients (HR 0.79, 95% CI 0.69 to 0.9, p<0.001). Following adjustments for baseline demographics and comorbidities, the relationship was no longer significant (HR 0.93, 95% CI 0.81 to 1.06, p=0.28 for 1-year mortality) (table 3).

    View this table:
    Table 3

    Multivariable Cox analysis

    Subgroup analyses and interaction analysis

    Subgroup analyses (figure 3) based on gender, diabetes and hypertension revealed that male, overweight (adjusted HR 0.8, 95% CI 0.68 to 0.96, p=0.014), and obese (adjusted HR 0.71, 95% CI 0.59 to 0.86, p<0.001) patients were strongly correlated with the 1-year mortality. We further assessed subgroups of diabetic (n=1022) and hypertensive (n=1301) patients, and revealed that, among these patient populations (diabetic (adjusted HR 0.72, 95% CI 0.56 to 0.93, p=0.012) and hypertensive (adjusted HR 0.73, 95% CI 0.58 to 0.92, p=0.008)), obesity was strongly related to enhanced OS. Furthermore, a strong interaction existed between BMI and mortality among diabetics (p=0.031) and hypertensive (p=0.035) patients.

    Figure 3
    Figure 3

    The association between BMI and 1-year mortality in subgroups. BMI, body mass index.

    Discussion

    This study identified 17 420 patients with sepsis. Among them, we selected 3145, based on our inclusion and exclusion criteria, and 1334 (42.4%) patients died within 1 year following the initial ICU admission. Table 1 summarises the cohort demographics, based on BMI status. Among all participants, 37.2% were obese and 31.0% were overweight. The 1-year mortality was reduced in overweight and obese patients, relative to normal-weight patients. When normal weight (BMI 18.5–24.9 kg/m2) was employed as a reference in Cox regression analysis, the obese patient 1-year mortality risk was 0.93 (95% CI 0.81 to 1.06, p=0.28). Diabetic (adjusted HR 0.72, 95% CI 0.56 to 0.93, p=0.012) and hypertensive (adjusted HR 0.73, 95% CI 0.58 to 0.92, p=0.008) patients were significantly related to the 1-year mortality in subgroup analysis. We also observed a marked interactive effect between BMI/diabetes/hypertension and the 1-year mortality. Based on our analyses, obese diabetic, hypertensive and male patients were at a reduced mortality risk at 1-year post-ICU admission, relative to normal weight patients (both unadjusted and adjusted analyses).

    Several prior studies investigated the association between BMI and sepsis-related mortality. However, the conclusion remains inconclusive.18 26–29 In a majority of studies, BMI was shown to be negatively correlated with sepsis-related mortality, and obesity greatly reduced the mortality risk of septic patients.6 18 27–29 Nevertheless, a retrospective observational investigation suggested that obese participants displayed a non-significant increase in mortality risk in adjusted 28-day mortality, relative to non-obese septic controls.26 A prospective study18 reported enhanced OS of morbidly obese patients (87%), relative to normal (84%) or underweight (78%) patients. However, this pattern did not reach significance, which corroborated with our results. One investigation revealed that augmented sagittal abdominal diameter, and not BMI, was markedly related to mortality among patients with sepsis. This indicated that BMI may ineffectively represent obesity.30 Another study investigated the association between BMI, muscle mass and OS in 10 140 Parkinson’ s disease patients. The aforementioned study concluded that patients with elevated BMI and enhanced or normal muscle mass experience the best OS.31 BMI poorly represents adiposity, and may not precisely reflect the fat status, muscle to fat ratio, or overall body composition.32 In the future, we recommended prospective investigations quantifying adipose tissue either at or before sepsis detection using anthropometric indexes of enhanced adiposity (eg, waist circumference, waist-to-hip ratio, percentage of visceral vs total body fat)33; or body composition evaluations via CT or dual-energy X-ray absorptiometry.

    We discovered gender differences in obesity and 1-year mortality; it is important to consider the clinical implications of this. One implication is the identification of at-risk populations. The average age of the women in this cohort was 66.2 years, implying that the majority were postmenopausal. In general, oestrogen has a protective effect on females. Because of the natural decrease in endogenous estradiol during menopause, postmenopausal women gain visceral fat mass, decreases muscle mass. This may account for the gender differences in the results.

    Adipose tissue was traditionally thought of as a storage site for lipids but is now recognised as an endocrine organ secreting signalling molecules and hormones that regulate metabolic homoeostasis. Adipocytokines are released by adipose tissue, which is made up of adipose cells and other immune cells. Adipocytokines that cause inflammation include ‘TNF- α’, ‘interleukin-6’ and ‘resistin’, etc. The most representative adipocytokine that inhibits inflammation is ‘adiponectin’.34

    Our analysis revealed a marked association between BMI and diabetes. Hypertensive19 and insulin resistant20 patients were also reported to experience enhanced septic patient OS, which corroborates well with our findings. This is likely due to alterations within the sympathetic nervous or endocrine system, which trigger a beneficial response in septic patients, or in the medications delivered to septic patients. Diabetes mellitus is strongly correlated with enhanced infectious disease-related mortality,35 yet, in another cohort,36 diabetes is strongly related to diminished mortality risk. Based on a retrospective investigation20 involving 792 patients with sepsis, obesity and morbid obesity were highly protective against patient death. Moreover, this protection may be due to the simultaneous presence of comorbid insulin resistance and diabetes, which is in agreement with our study. In another retrospective study, a U-shaped relationship was observed between BMI and all-cause mortality in type 2 diabetic patients, who carried the lowest mortality risk among all overweight patients.37 Among hypertensive CAD patients, overweight or obesity offered more protection against patient death or development of nonfatal myocardial infarction or stroke.19 It is well known that obese hypertensive patients tend to exhibit an elevated cardiac output, excess blood volume, and reduced systemic vascular resistance, relative to normal weight patients.38–40 Since systemic resistance represents the hypertensive cardiovascular disease status, the relatively reduced values in obese patients may translate to an enhanced prognosis within this patient population. There may be multiple reasons behind the observed obesity paradox. First, obese patients tend to experience more frequent diabetes check-ups, which may result in the early detection and better management of the disease, thereby producing enhanced outcomes. As has been reported in prior investigations,41 42 the mean age of obese participants was considerably younger than the normal-weight category in our study. Therefore, overweight individuals may possess a metabolic reserve that protects them from adverse or undesirable outcomes.

    We also made adjustments for the age, gender, disease, severity, vital signs and comorbidities variables to determine whether these factors affect patient outcome. Based on our analyses, overweight/obese patients experienced better OS, however, it did not reach significance. Another explanation may be that patients with type 2 diabetes, owing to their obesity-related metabolic stress, may have enhanced prognosis and reduced complications and comorbidity risk,43 relative to a low BMI genetic susceptibility-induced type 1 diabetic.44 Additional investigations are needed to delineate between the effects of type 1 and type 2 diabetes mellitus. Diabetic and hypertensive patients, with poor control or advanced underlying disease, may lose weight unintentionally and experience reduced or normal BMI. This can enhance mortality risk by incorrectly suggesting that augmented BMI protects against patient death in diabetic and hypertensive patients.

    Our work has the following limitations: First, the research was retrospective. Therefore, unlike other observational investigations, there may be some unintentional biases that influenced the results. Second, the MIMIC database used in this study had missing information, which in some indexes was as high as 28%. This may have introduced potential selection bias. We have included in this manuscript our method of dealing with missing height records, which we confirmed was unbiased. Also, when conducting the final regression analysis, any baseline variables with over 5% of missing values were excluded from analysis. Third, our findings invoke discussion and hypotheses regarding the topic of BMI and mortality link, but do not imply causality. It is possible that some confounding factors remained in our quantitative data, particularly, due to the exclusion of smoking status42 and diabetes type in our multivariate analysis. Lastly, we measured adiposity using BMI. Hence, there was no direct assessment of body composition, such as, visceral fat or fat distribution.

    Conclusions

    In conclusion, according to our analysis, men and and obese patients with diabetes and hypertension have a reduced risk of sepsis-related mortality compared with normal weight patients. This essential information is highly beneficial to septic patient intervention and clinical prognosis. Combining the findings of BMI and mortality risk, recommendations for body weight management should be individualised for different patients. Our conclusion requires further validation in a prospective investigation.

    Data availability statement

    Data are available on reasonable request. Data for this study were obtained from MIMIC III, which was approved by Massachusetts Institute of Technology (No. 0403000206) and Beth Israel Deaconess Medical Center (2001-P-001699/14).

    Ethics statements

    Patient consent for publication

    Ethics approval

    After successfully completing the National Institutes of Health Protecting Human Research Participants training course, We obtained approval to use MIMIC-III data for research (Certification Number: 38807989).

    References

    1. ↵
      1. Rhee C,
      2. Dantes R,
      3. Epstein L, et al
      . Incidence and trends of sepsis in US hospitals using clinical vs claims data, 2009-2014. JAMA 2017;318:1241–9. doi:10.1001/jama.2017.13836
      OpenUrlCrossRefPubMed
    2. ↵
      1. Kaukonen K-M,
      2. Bailey M,
      3. Suzuki S, et al
      . Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA 2014;311:1308–16. doi:10.1001/jama.2014.2637
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Angus DC,
      2. van der Poll T
      . Severe sepsis and septic shock. N Engl J Med 2013;369:840–51. doi:10.1056/NEJMra1208623
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Gaieski DF,
      2. Edwards JM,
      3. Kallan MJ, et al
      . Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med 2013;41:1167–74. doi:10.1097/CCM.0b013e31827c09f8
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Shankar-Hari M,
      2. Harrison DA,
      3. Rubenfeld GD, et al
      . Epidemiology of sepsis and septic shock in critical care units: comparison between sepsis-2 and sepsis-3 populations using a national critical care database. Br J Anaesth 2017;119:626–36. doi:10.1093/bja/aex234
      OpenUrlPubMed
    6. ↵
      1. Pepper DJ,
      2. Demirkale CY,
      3. Sun J, et al
      . Does obesity protect against death in sepsis? A retrospective cohort study of 55,038 adult patients. Crit Care Med 2019;47:643–50. doi:10.1097/CCM.0000000000003692
      OpenUrlPubMed
    7. ↵
      1. Hogue CW,
      2. Stearns JD,
      3. Colantuoni E, et al
      . The impact of obesity on outcomes after critical illness: a meta-analysis. Intensive Care Med 2009;35:1152–70. doi:10.1007/s00134-009-1424-5
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Akinnusi ME,
      2. Pineda LA,
      3. El Solh AA
      . Effect of obesity on intensive care morbidity and mortality: a meta-analysis. Crit Care Med 2008;36:151–8. doi:10.1097/01.CCM.0000297885.60037.6E
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Karampela I,
      2. Chrysanthopoulou E,
      3. Christodoulatos GS, et al
      . Is there an obesity paradox in critical illness? epidemiologic and metabolic considerations. Curr Obes Rep 2020;9:231–44. doi:10.1007/s13679-020-00394-x
      OpenUrl
    10. ↵
      1. Endalifer ML,
      2. Diress G
      . Epidemiology, predisposing factors, biomarkers, and prevention mechanism of obesity: A systematic review. J Obes 2020;2020:6134362. doi:10.1155/2020/6134362
    11. ↵
      1. Mewes C,
      2. Böhnke C,
      3. Alexander T, et al
      . Favorable 90-day mortality in obese Caucasian patients with septic shock according to the sepsis-3 definition. J Clin Med 2019;9:46. doi:10.3390/jcm9010046
    12. ↵
      1. Mei X,
      2. Hu S,
      3. Mi L, et al
      . Body mass index and all-cause mortality in patients with percutaneous coronary intervention: a dose-response meta-analysis of obesity paradox. Obes Rev 2021;22:e13107. doi:10.1111/obr.13107
    13. ↵
      1. Park J,
      2. Ahmadi S-F,
      3. Streja E, et al
      . Obesity paradox in end-stage kidney disease patients. Prog Cardiovasc Dis 2014;56:415–25. doi:10.1016/j.pcad.2013.10.005
      OpenUrlCrossRefPubMed
    14. ↵
      1. Ng PY,
      2. Eikermann M
      . The obesity conundrum in sepsis. BMC Anesthesiol 2017;17:147. doi:10.1186/s12871-017-0434-z
    15. ↵
      1. Marques MB,
      2. Langouche L
      . Endocrine, metabolic, and morphologic alterations of adipose tissue during critical illness. Crit Care Med 2013;41:317–25. doi:10.1097/CCM.0b013e318265f21c
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Kershaw EE,
      2. Flier JS
      . Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 2004;89:2548–56. doi:10.1210/jc.2004-0395
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Winkler G,
      2. Kiss S,
      3. Keszthelyi L, et al
      . Expression of tumor necrosis factor (TNF) -alpha protein in the subcutaneous and visceral adipose tissue in correlation with adipocyte cell volume, serum TNF-alpha, soluble serum TNF-receptor-2 concentrations and C-peptide level. Eur J Endocrinol 2003;149:129–35. doi:10.1530/eje.0.1490129
      OpenUrlAbstract
    18. ↵
      1. Dossett LA,
      2. Dageforde LA,
      3. Swenson BR, et al
      . Obesity and site-specific nosocomial infection risk in the intensive care unit. Surg Infect (Larchmt) 2009;10:137–42. doi:10.1089/sur.2008.028
      OpenUrlCrossRefPubMed
    19. ↵
      1. Uretsky S,
      2. Messerli FH,
      3. Bangalore S, et al
      . Obesity paradox in patients with hypertension and coronary artery disease. Am J Med 2007;120:863–70. doi:10.1016/j.amjmed.2007.05.011
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Kuperman EF,
      2. Showalter JW,
      3. Lehman EB, et al
      . The impact of obesity on sepsis mortality: a retrospective review. BMC Infect Dis 2013;13:377. doi:10.1186/1471-2334-13-377
    21. ↵
      1. Chang E,
      2. Varghese M,
      3. Singer K
      . Gender and sex differences in adipose tissue. Curr Diab Rep 2018;18:69. doi:10.1007/s11892-018-1031-3
    22. ↵
      1. Angus DC,
      2. Linde-Zwirble WT,
      3. Lidicker J, et al
      . Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303–10. doi:10.1097/00003246-200107000-00002
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Vincent JL,
      2. Moreno R,
      3. Takala J, et al
      . The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure. on behalf of the Working group on sepsis-related problems of the European Society of intensive care medicine. Intensive Care Med 1996;22:707–10. doi:10.1007/BF01709751
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. van Walraven C,
      2. Austin PC,
      3. Jennings A, et al
      . A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care 2009;47:626–33. doi:10.1097/MLR.0b013e31819432e5
      OpenUrlCrossRefPubMed
    25. ↵
      1. Le Gall JR,
      2. Loirat P,
      3. Alperovitch A, et al
      . A simplified acute physiology score for ICU patients. Crit Care Med 1984;12:975–7. doi:10.1097/00003246-198411000-00012
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Gaulton TG,
      2. Weiner MG,
      3. Morales KH, et al
      . The effect of obesity on clinical outcomes in presumed sepsis: a retrospective cohort study. Intern Emerg Med 2014;9:213–21. doi:10.1007/s11739-013-1002-2
      OpenUrlCrossRefPubMed
    27. ↵
      1. Li Q,
      2. Tong Y,
      3. Liu S, et al
      . Association between body mass index and short-term mortality in patients with intra-abdominal infections: a retrospective, single-centre cohort study using the medical information Mart for intensive care database. BMJ Open 2021;11:e046623. doi:10.1136/bmjopen-2020-046623
    28. ↵
      1. Li S,
      2. Hu X,
      3. Xu J, et al
      . Increased body mass index linked to greater short- and long-term survival in sepsis patients: a retrospective analysis of a large clinical database. Int J Infect Dis 2019;87:109–16. doi:10.1016/j.ijid.2019.07.018
      OpenUrlPubMed
    29. ↵
      1. Wurzinger B,
      2. Dünser MW,
      3. Wohlmuth C, et al
      . The association between body-mass index and patient outcome in septic shock: a retrospective cohort study. Wien Klin Wochenschr 2010;122:31–6. doi:10.1007/s00508-009-1241-4
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Chalkias A,
      2. Nitsotolis T,
      3. Papalexandrou A, et al
      . Sagittal abdominal diameter may effectively predict future complications and increased mortality in intensive care unit patients with severe sepsis. J Crit Care 2013;28:964–9. doi:10.1016/j.jcrc.2013.06.022
      OpenUrl
    31. ↵
      1. Ramkumar N,
      2. Pappas LM,
      3. Beddhu S
      . Effect of body size and body composition on survival in peritoneal dialysis patients. Perit Dial Int 2005;25:461–9. doi:10.1177/089686080502500510
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Christopher KB
      . The body mass index paradox. Crit Care Med 2015;43:2693–4. doi:10.1097/CCM.0000000000001344
      OpenUrl
    33. ↵
      1. Pichard C,
      2. Kyle UG,
      3. Morabia A, et al
      . Nutritional assessment: lean body mass depletion at hospital admission is associated with an increased length of stay. Am J Clin Nutr 2004;79:613–8. doi:10.1093/ajcn/79.4.613
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Maximus PS,
      2. Al Achkar Z,
      3. Hamid PF, et al
      . Adipocytokines: are they the theory of everything? Cytokine 2020;133. doi:10.1016/j.cyto.2020.155144
    35. ↵
      1. Zoppini G,
      2. Fedeli U,
      3. Schievano E, et al
      . Mortality from infectious diseases in diabetes. Nutr Metab Cardiovasc Dis 2018;28:444–50. doi:10.1016/j.numecd.2017.12.007
      OpenUrlPubMed
    36. ↵
      1. Lin C-C,
      2. Li C-I,
      3. Liu C-S, et al
      . Obesity paradox in associations between body mass index and diabetes-related hospitalization and mortality in patients with type 2 diabetes: retrospective cohort studies. Diabetes Metab 2019;45:564–72. doi:10.1016/j.diabet.2019.02.007
      OpenUrl
    37. ↵
      1. Salehidoost R,
      2. Mansouri A,
      3. Amini M, et al
      . Body mass index and the all-cause mortality rate in patients with type 2 diabetes mellitus. Acta Diabetol 2018;55:569–77. doi:10.1007/s00592-018-1126-y
      OpenUrl
    38. ↵
      1. Barrett-Connor E,
      2. Khaw KT
      . Is hypertension more benign when associated with obesity? Circulation 1985;72:53–60. doi:10.1161/01.cir.72.1.53
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Messerli FH
      . Cardiovascular adaptations to obesity and arterial hypertension: detrimental or beneficial? Int J Cardiol 1983;3:94–7. doi:10.1016/0167-5273(83)90069-4
      OpenUrlPubMed
    40. ↵
      1. Messerli FH,
      2. Sundgaard-Riise K,
      3. Reisin E, et al
      . Disparate cardiovascular effects of obesity and arterial hypertension. Am J Med 1983;74:808–12. doi:10.1016/0002-9343(83)91071-9
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Costanzo P,
      2. Cleland JGF,
      3. Pellicori P, et al
      . The obesity paradox in type 2 diabetes mellitus: relationship of body mass index to prognosis: a cohort study. Ann Intern Med 2015;162:610–8. doi:10.7326/M14-1551
      OpenUrlCrossRefPubMed
    42. ↵
      1. Tobias DK,
      2. Pan A,
      3. Jackson CL, et al
      . Body-Mass index and mortality among adults with incident type 2 diabetes. N Engl J Med 2014;370:233–44. doi:10.1056/NEJMoa1304501
      OpenUrlCrossRefPubMedWeb of Science
    43. ↵
      1. Perry JRB,
      2. Voight BF,
      3. Yengo L, et al
      . Stratifying type 2 diabetes cases by BMI identifies genetic risk variants in Lama1 and enrichment for risk variants in lean compared to obese cases. PLoS Genet 2012;8:e1002741. doi:10.1371/journal.pgen.1002741
    44. ↵
      1. Wing RR,
      2. Lang W,
      3. Wadden TA, et al
      . Benefits of modest weight loss in improving cardiovascular risk factors in overweight and obese individuals with type 2 diabetes. Diabetes Care 2011;34:1481–6. doi:10.2337/dc10-2415
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors LZ and LF were equally responsible for the writing of the manuscript. LZ conducted the statistical analyses. CL and HL participated in the study design and conduct and assisted in the writing of the manuscript. LZ act as guarantor.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.