Разнообразие по признаку пола 90-Day Послеоперационные осложнения у пациентов с анемической атропластикой.

Introduction

Total hip arthroplasty (THA) is one of the most frequently performed orthopaedic procedures, and as populations age, optimizing perioperative risk factors increases in importance [ 1 ]. Preoperative anemia, defined by the World Health Organization (WHO) as hemoglobin <12 g/dL in females and <13 g/dL in males [ 2 ], affects a substantial portion of THA candidates, particularly females [ 3 ], and is independently associated with transfusion, wound complications, infection, organ dysfunction, prolonged hospitalization, and mortality [ [4] , [5] , [6] , [7] , [8] ]. THA is also associated with greater average blood loss than many other elective surgeries, making anemia a particularly impactful risk factor [ 9 , 10 ].

Postoperative complications also vary by sex, with females experiencing higher rates of urinary tract infection (UTI), transfusion, and readmission while males incur more acute kidney injury (AKI) and surgical site infections [ 11 , 12 ]. Although estrogen has been associated with protective effects against ischemia-reperfusion injury, females generally have lower baseline hemoglobin, iron reserves, and circulating blood volume, which may contribute to differences in preoperative anemia and transfusion requirements [ 8 , 13 ]. This potentially increases the risk for adverse outcomes when compounded by the higher average blood loss seen in orthopaedic surgery. Some studies of nonsurgical populations suggest that it is males that may experience a greater increase in mortality or morbidity when anemic [ 14 , 15 ], while one study of noncardiac surgeries found no sex differences among anemic patients [ 16 ].

Guidelines acknowledge anemia as a general surgical risk but do not address its interaction with sex or offer sex-specific recommendations [ 17 ]. With anemia and biological sex as established risk factors, few investigations have examined their combined effect. To fill this gap, we performed a retrospective propensity-matched analysis of a large multi-institutional database to compare postoperative complication profiles between male and female patients with preoperative anemia undergoing THA. We hypothesized that anemic females would demonstrate higher risks of wound-related and readmission complications due to lower baseline hemoglobin and iron reserves, whereas anemic males would show greater risks of thromboembolic and renal complications.

Material and methods

TriNetX Research Network

This retrospective cohort study used the TriNetX Research Network (Cambridge, MA), a federated platform providing deidentified electronic health records from 107 healthcare organizations as of August 2025. No external funding was required. Procedural and diagnostic data were extracted using International Classification of Diseases, Tenth Revision (ICD-10) and Current Procedural Terminology (CPT) codes. Institutional review board approval was not required for this study because TriNetX provides fully deidentified data in compliance with the Health Insurance Portability and Accountability Act regulations.

Defining the cohorts

Patients ≥18 years who underwent primary THA (CPT 27130) and had documented anemia within 3 months preoperatively were included. Anemia was defined by WHO thresholds and adapted to TriNetX: hemoglobin ≤12.99 g/dL for males and ≤11.99 g/dL for females. Hemoglobin values were identified using the TriNetX laboratory data element “Hemoglobin in blood (g/dL)” within a 3-month preoperative window. Patients were included if any recorded value met the predefined anemia thresholds and divided into 2 cohorts based on biological sex as defined within the database. Data used only included 5 years prior to the date of access.

Outcomes of interest

Postoperative complications were assessed from day 1 postsurgery to day 90. Outcomes included infections (prosthetic joint infection [PJI], sepsis, pneumonia, UTI), thromboembolic events (deep vein thrombosis [DVT], pulmonary embolism [PE]), wound dehiscence, hematologic events such as transfusion, AKI, readmission, and mortality. Outcomes were defined by ICD-10 and CPT codes ( Supplemental Table 1 ) and were captured from any documented encounter within participating healthcare organizations in the TriNetX network. Except for readmission/hospitalization, which was defined as an inpatient encounter, postoperative complications were not restricted to inpatient-only coding.

Propensity score matching

To reduce confounding, propensity scores were estimated using logistic regression within the TriNetX analytics platform, and anemic male and anemic female patients were then matched 1:1. This approach results in equal cohort sizes by design. Matching variables included age, race, body mass index, smoking status, primary/secondary hypertension, rheumatoid arthritis, osteoporosis, frailty, socioeconomic status, and the Charlson Comorbidity Index (CCI). Socioeconomic status was approximated using documented ICD-10 codes Z55-Z65, which capture social determinants of health and psychosocial circumstances, including factors related to education, employment, housing, family circumstances, and other socioeconomic or environmental determinants. We adapted our CCI methodology to match a previously published TriNetX-based approach [ 18 ]. To enhance diagnostic accuracy, additional ICD-10 codes capturing diabetes subtypes were included beyond the standard CCI definitions. Several ICD-10 codes included in the original CCI were not available within the TriNetX Research Network and therefore could not be incorporated; these omissions are listed in Supplemental Table 2 . A complete list of ICD-10 codes used to define study outcomes and covariates is provided in Table 1 , Supplemental Tables 1 and 2 . A successful match was achieved with a standardized mean difference <0.1 for all measured variables, all of which are outlined in Supplemental Table 3 .

Table 1 Additional non-Charlson Comorbidity Index covariates included in propensity score matching. Controls ICD-10 codes Primary/secondary hypertension I10, I15 Osteoporosis M80, M81 Rheumatoid arthritis M05, M06 Frailty R54 Socioeconomic status Z55-Z65 Tobacco use Z72.0 The adapted Charlson Comorbidity Index was also included in propensity score matching and is detailed in Supplemental Table 2 .

Additional non-Charlson Comorbidity Index covariates included in propensity score matching.

The adapted Charlson Comorbidity Index was also included in propensity score matching and is detailed in Supplemental Table 2 .

Statistical analysis

In the matched cohort, we compared complication rates using odds ratios (ORs) and 95% confidence intervals (CIs), with statistical significance defined as P < .05. Time-to-event outcomes were analyzed with Kaplan-Meier curves and log-rank tests, and hazard ratios (HRs) were estimated via Cox proportional hazards models. All analyses used TriNetX’s built-in analytics. Figure 1 below demonstrates the propensity score matching where standardized mean difference for all covariates was <0.1.

Propensity score matching demonstrated through visual representation of anemic males in purple and anemic females in green.

Results

Following propensity score matching of the original anemic male (n = 27,402) and anemic female (n = 36,326) cohorts, the final analytic sample comprised 22,894 patients in each group. Sex-based 90-day postoperative complication rates with corresponding ORs, HRs, and P values are summarized in Table 2 and visually represented in Figure 2 .

Table 2 Odds ratios and hazard ratios with 95% confidence intervals for 90-day postoperative complications following total hip arthroplasty among anemic patients, stratified by sex. Complication Odds ratio 95% confidence interval Hazard ratio 95% confidence interval P value Favored sex Acute kidney injury 1.15 1.05-1.27 1.16 1.06-1.27 .0030 Male Blood transfusion 0.63 0.57-0.71 0.64 0.57-0.71 <.0001 Female Deep vein thrombosis 1.20 1.02-1.41 1.20 1.02-1.41 .0298 Male Mortality 1.22 0.99-1.51 1.23 1.00-1.52 .0575 No significant difference Prosthetic joint infection 1.03 0.82-1.30 1.04 0.83-1.31 .7717 No significant difference Pneumonia 0.98 0.84-1.13 0.98 0.85-1.14 .7358 No significant difference Pulmonary embolism 1.04 0.87-1.23 1.04 0.88-1.23 .6960 No significant difference Readmission/hospitalization 0.93 0.88-0.97 0.94 0.90-0.98 .0015 Female Sepsis 1.01 0.86-1.20 1.02 0.86-1.21 .8977 No significant difference Urinary tract infection 0.53 0.47-0.59 0.54 0.48-0.60 <.0001 Female Wound dehiscence 0.58 0.48-0.69 0.59 0.49-0.70 <.0001 Female Odds ratio from logistic regression; hazard ratio from Cox proportional hazards model (Kaplan–Meier/log-rank). All confidence intervals are 95%.

Odds ratios and hazard ratios with 95% confidence intervals for 90-day postoperative complications following total hip arthroplasty among anemic patients, stratified by sex.

Odds ratio from logistic regression; hazard ratio from Cox proportional hazards model (Kaplan–Meier/log-rank). All confidence intervals are 95%.

Forest plot demonstrating odds ratios with 95% confidence intervals for 90-day postoperative complications following THA, stratified by sex. Odds ratios are plotted relative to the null value (OR = 1.0, dashed vertical line). Complications with statistically significant differences favoring females are shown in red, those favoring males are shown in blue, and outcomes without statistical significance are shown in black.

Readmission or hospitalization

Within 90 days postsurgery, 19.0% of anemic females experienced readmission or hospitalization compared with 17.9% of anemic males ( P = .0015; OR = 0.93; 95% CI: 0.88-0.97; HR = 0.94; 95% CI: 0.90-0.98). Kaplan-Meier analysis demonstrated a significantly lower hazard of readmission among males, consistent with the overall OR findings ( Table 2 , Fig. 2 ). Among patients who were readmitted, the mean number of readmissions did not differ significantly between sexes (2.33 ± 0.04 in males vs 2.26 ± 0.04 in females; P = .2705). Readmission and hospitalization outcomes reflect all-cause inpatient encounters within 90 days as defined by the platform, as specific reasons for readmission are not uniformly available in TriNetX.

Wound dehiscence

Within 90 days postsurgery, wound dehiscence occurred in 1.4 % of anemic females compared with 0.8% of anemic males ( P < .0001; OR = 0.58; 95% CI: 0.48-0.69; HR = 0.59; 95% CI: 0.49-0.70). Kaplan-Meier analysis demonstrated a significantly lower hazard of dehiscence among males ( Table 2 , Fig. 2 ). The 90-day wound-dehiscence-free survival probability was 99.1% for males and 98.5% for females. Among those who experienced dehiscence, the mean number of events per patient did not differ significantly between sexes (2.08 ± 0.09 in males vs 2.19 ± 0.08 in females; P = .5704).

Infectious complications

Several complications showed no statistically significant difference between anemic males and females undergoing THA, as shown in Table 2 . Sepsis occurred in 1.2% of males vs 1.2% of females ( P = .8977; OR = 1.011; 95% CI: 0.855-1.196; HR = 1.02; 95% CI: 0.9-1.2). PJI rates were similarly low (0.66% in males vs 0.64% in females; P = .7717; OR = 1.034; 95% CI: 0.8–1.3; HR = 1.043; 95% CI: 0.832-1.309). Pneumonia occurred at comparable rates as well (1.56% in males vs 1.60% in females; P = .7358; OR = 0.975; 95% CI: 0.8-1.1; HR = 0.983; 95% CI: 0.8-1.1).

Thromboembolic complications

PE occurred in 1.2% of anemic males and 1.2% of anemic females ( P = .6960; OR = 1.04; 95% CI: 0.87-1.23; HR = 1.04; 95% CI: 0.88-1.23). DVT occurred in 1.4% of anemic males vs 1.2% of anemic females ( P = .0298; OR = 1.20; 95% CI: 1.02-1.41; HR = 1.20; 95% CI: 1.02-1.41) ( Table 2 , Fig. 2 ).

Systemic and hematologic complications

Mortality demonstrated no significant difference but trended toward males, occurring in 0.9% of anemic males vs 0.7% of anemic females ( P = .0575; OR = 1.22; 95% CI: 0.99-1.51; HR = 1.23; 95% CI: 1.0-1.5). Blood transfusions demonstrated a female predominance, occurring in 3.7% of anemic females vs 2.4% of anemic males ( P < .0001; OR = 0.63; 95% CI: 0.58-0.71; HR = 0.64; 95% CI: 0.57-0.71) ( Table 2 , Fig. 2 ).

Genitourinary complications

Among anemic patients undergoing THA, 4.3% of males and 3.8% of females experienced AKI ( P = .0030; OR = 1.15; 95% CI: 1.05-1.27; HR = 1.16; 95% CI: 1.06-1.27). In contrast, UTI occurred in 1.9% of anemic males vs 3.6% of anemic females ( P < .0001; OR = 0.53; 95% CI: 0.47-0.59; HR = 0.54; 95% CI: 0.48-0.60) ( Table 2 , Fig. 2 ).

Discussion

This study evaluated sex-based differences in 90-day postoperative complications among anemic patients undergoing THA. While anemia and sex independently affect surgical outcomes, few studies examine their potential interaction.

Readmission and hospitalization

Readmission and hospitalization following THA have historically been more common in males and independently associated with anemia [ 4 , 12 , 19 , 20 ]. However, in our matched anemic cohort, females experienced significantly higher readmission and hospitalization rates within 90 days. Kaplan-Meier analysis further confirmed this trend, showing a lower readmission-free survival probability among females. These results may reflect sex-related differences in postoperative risk among anemic patients, though the underlying mechanisms cannot be determined from this dataset. Given the higher prevalence of anemia among females, and the possibility of sex-related differences in physiologic responses described in prior literature, these findings support further evaluation of individualized preoperative optimization strategies [ 21 ]. WHO anemia cutoffs separated by sex apply to adults 15-65 and offer no guidance for older patients. Our female cohort had a mean age of 65.5 ± 13 years after matching. This raises the question of whether age-adjusted or more nuanced anemia thresholds are needed.

Wound dehiscence

Wound dehiscence occurred more often in anemic females, contradicting prior studies linking wound complications to male sex [ 12 , 22 ]. Prior literature has described associations between low hemoglobin levels and impaired oxygen delivery or collagen synthesis; however, such mechanisms cannot be evaluated within this dataset, and the basis of this association remains uncertain [ 21 ]. Although prior studies have suggested that hormonal factors may influence tissue repair, such mechanisms also cannot be assessed in this dataset and were not directly evaluated in our analysis [ 23 , 24 ].

Infectious complications

Sepsis, PJI, and pneumonia demonstrated no statistically significant sex-based differences among anemic patients undergoing THA. These complications are frequently reported in the literature as occurring more commonly in males following joint arthroplasty [ 12 , 22 , 25 , 26 ], and each has also been independently associated with preoperative anemia [ 8 , [27] , [28] , [29] ]. Although females generally have lower baseline hemoglobin, circulating blood volume, and iron stores than males, which may influence physiologic responses to anemia, our study did not demonstrate sex-based differences in postoperative infection [ [30] , [31] , [32] ]. It should be noted this study examined patients retrospectively over 5 years, thus the COVID-19 pandemic may have influenced the rates of pneumonia present.

Thromboembolic complications

PE rates did not differ significantly between anemic males and females, whereas DVT rates showcased higher rates among anemic males. Prior literature reports a slight increase in venous thromboembolism (VTE) risk among females in the general THA population, although these analyses typically assess VTE as a composite outcome rather than evaluating DVT and PE separately [ [33] , [34] , [35] ]. Anemia alone has not been shown to increase VTE risk, with transfusion identified as the relevant driver [ 36 , 37 ]. A recent study did find male sex as an independent risk factor for VTE following THA [ 38 ], aligning with the higher DVT rate observed in our cohort. Notably, the DVT difference in our cohort reached statistical significance ( P = .0298) but was modest in magnitude, suggesting this association should be interpreted with caution.

Systemic and hematologic complications

Mortality showed no difference between anemic males and females, contrasting with prior literature identifying male sex [ 12 , 22 , 39 ], and preoperative anemia [ 4 , 8 , 40 , 41 ] as independent risk factors. There was a trend toward male sex, shown by the P value of 0.0575 and a difference of 0.16%. Females generally possess lower baseline hemoglobin concentrations, which may influence physiologic reserve, although our data did not demonstrate sex-based differences in mortality [ 3 ]. Females also have smaller circulating blood volumes and lower total hemoglobin mass than males, factors that have been proposed in prior literature to affect physiologic responses to anemia; however, such mechanisms cannot be evaluated within this dataset [ [42] , [43] , [44] ]. These findings emphasize the current call to examine anemia cutoffs.

Transfusion rates were also significantly higher in females, aligning with prior literature [ 10 , [45] , [46] , [47] ]. Despite higher transfusion rates in females, we observed no increase in VTE risk in our cohort; however, as previously discussed, there was an observed higher rate of DVT in anemic males. This contrasts with prior data and may reflect variable prophylaxis or differences in how transfusion-related risk manifests in anemic patients and warrants investigation.

Genitourinary complications

Genitourinary complications, namely AKI and UTI, aligned with established sex-based trends in the literature, and our findings show that these associations persist within the anemic population following THA. AKI has been independently linked to both preoperative anemia [ 4 , 5 , 8 ] and male sex [ 22 , 48 ], and we observed higher rates of AKI among anemic males. Conversely, UTIs are more commonly associated with preoperative anemia [ 4 , 5 , 8 ] and female sex [ 49 ], a trend mirrored by our finding of higher UTI rates in anemic females.

Clinical practice

The observed attenuation and elevation of postoperative complications underscore the need for continued investigation into preoperative risk assessment in anemic patients undergoing THA. Sex-based comparison revealed multiple divergences from prior literature, emphasizing the importance of further sex-disaggregated investigation. Anemia, as a high-risk preoperative condition, warrants closer evaluation to refine risk definitions and thresholds. Current practice separates anemia cutoffs by sex; however, our findings indicate that refinement of sex-specific risk profiles may be warranted. These findings highlight the importance for arthroplasty teams of integrating sex-specific risk estimates into perioperative planning. Proactive optimization of anemic patients within THA strategies such as minimizing blood loss, maintaining vigilance for complications, and ensuring transparent, sex-specific counseling may warrant consideration in future clinical workflows and support informed shared decision-making within this surgical population.

Limitations

While this study addresses a previously unexplored question in postoperative complications among anemic patients undergoing THA, several limitations must be acknowledged. As a retrospective analysis using aggregated electronic health record data from TriNetX, the study lacks patient-specific detail. Data were identified through ICD-10 and CPT codes, which are subject to misclassification. Furthermore, several ICD-10 codes included in the CCI were not available within the TriNetX database and are listed in Supplemental Table 2 . Given the rarity of these codes and their limited relevance to a primary THA population, their exclusion is unlikely to meaningfully affect cohort characterization.

Additionally, indication for THA (eg, elective osteoarthritis vs fracture-related arthroplasty) could not be reliably stratified and may differ by sex, representing a potential source of residual confounding. Surgical approach, implant type, operative time, anesthesia type, perioperative blood management (including tranexamic acid use), thromboprophylaxis, and discharge disposition are not captured within TriNetX, limiting adjustment for perioperative management differences. Propensity score matching was used to balance cohorts, but residual confounding remains possible. Data were drawn from over 100 healthcare organizations with variable protocols, introducing institutional and temporal heterogeneity.

As this study relied on deidentified, structured electronic health record data across multiple healthcare organizations, patient-level rates of missing data for individual covariates were not available for direct reporting, and no imputation was performed.

Although hemoglobin values were available, anemia was defined categorically using WHO thresholds. Due to limitations of the TriNetX platform, a single standardized preoperative hemoglobin value per patient could not be consistently selected, and stratification by anemia severity (eg, mild, moderate, severe) was not performed. Since hemoglobin values were used to define anemia status, postmatch hemoglobin distributions were not included as matching covariates, limiting assessment of whether differences in anemia severity or hemoglobin concentration persisted between matched cohorts. Because anemia status was assigned if any hemoglobin value within the 3-month preoperative window met WHO thresholds, some patients may have undergone THA after partial or complete correction of anemia, which could bias associations toward the null. Additionally, because continuous hemoglobin levels, nadir perioperative values, and degree of deviation from baseline could not be analyzed, we were unable to assess dose-response relationships or sex-specific threshold effects.

Readmission outcomes reflect all-cause hospitalizations, as reasons for readmission are not uniformly available. Events occurring outside participating healthcare systems may not be captured within TriNetX, introducing the potential for underestimation of complication rates.

The study period included the COVID-19 pandemic, which may have influenced pneumonia diagnoses, hospitalization patterns, and coding practices, contributing additional temporal bias. Nonetheless, this study benefits from a large, matched sample and includes analysis of both common and rare complications in a high-risk group.

Conclusions

Among anemic patients undergoing THA, a matched sex-based comparison revealed both expected and unexpected postoperative trends. Anemic males had higher rates of DVT and AKI, while anemic females showed increased rates of readmission, wound dehiscence, and UTI. No significant sex-based differences were observed in sepsis, PJI, PE, pneumonia, or mortality despite prior literature suggesting male predominance. These findings highlight the need to examine sex-specific risk within high-risk surgical cohorts. Further research is needed to validate these results and guide more targeted perioperative management.

CRediT authorship contribution statement

Brendan Layton: Writing – review & editing, Writing – original draft, Visualization, Validation, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Dilan Prasad: Writing – review & editing, Methodology, Formal analysis, Data curation, Conceptualization. Dhara Patel: Writing – review & editing, Validation, Supervision, Project administration, Conceptualization. Saqib Rehman: Writing – review & editing, Validation, Supervision, Project administration, Formal analysis, Conceptualization.

Conflicts of interest

S. Rehman receives royalties from Globus Medical and is a member of medical/orthopaedic publications editorial/governing board at SurgiColl and Orthopedics Today; all other authors declare no potential conflicts of interest.

For full disclosure statements refer to https://doi.org/10.1016/j.artd.2026.102068 .

Supplementary data

Conflict of Interest Statement for Layton

Conflict of Interest Statement for Rehman

Conflict of Interest Statement for Prasad

Conflict of Interest Statement for Patel

Appendix

.

Supplemental Table 1 ICD-10 codes used for postoperative complications Complication ICD-10 codes Prosthetic joint infection ICD-10-CM T84.5 Mortality Term description: Deceased Readmission Term description: Visit: Inpatient Encounter Sepsis ICD-10-CM A41.9 Transfusion CPT 36455; CPT 36430 Deep vein thrombosis ICD-10-CM I82.4 Pulmonary embolism ICD-10-CM I26 Wound dehiscence ICD-10-CM T81.3 Acute kidney injury ICD-10-CM N17 Pneumonia ICD-10-CM J18 Urinary tract infection ICD-10-CM N39.0

ICD-10 codes used for postoperative complications

Supplemental Table 2 ICD-10 codes used for adaptation of Charlson Comorbidity Index Comorbidity ICD-10 codes Acquired immunodeficiency syndrome B20.x Cerebrovascular disease G45.x, G46.x, H34.0, I60.x-I69.x Chronic pulmonary disease I27.8, I27.9, J40.x-J45.x, J47.x, J60.x-J67.x, J68.4, J70.1, J70.3 Congestive heart failure I09.9, I11.0, I13.0, I13.2, I25.5, I42.0, I42.5-I42.9, I43.x, I50.x Connective tissue disease M05.x, M06.x, M31.5, M32.x-M34.x, M35.1, M35.3, M36.0 Diabetes E08.x-E11.x, E13.x Liver disease B18.x, K70.x-K77.x, Z94.4, I85.0, I86.4 Myocardial infarction I21.x, I22.x, I25.2, I09.9, I11.0, I13.0, I13.2, I25.5, I42.0, I42.5-I42.9, I43.x, I50.x Peptic ulcer disease K25.x-K28.x Peripheral vascular disease I70.x, I71.x, I73.1, I73.8, I73.9, I77.1, I79.0, K55.1, K55.8, K55.9, Z95.8, Z95.9 Plegia G04.1, G11.4, G80.1, G80.2, G81.x, G82.x, G83.0-G83.4, G83.9 Renal disease I12.0, I13.1, N03.2-N03.7, N05.2-N05.7, N18.x, N19.x, N25.0, Z49.0, Z99.2 ICD-10 codes not available in TriNetX (excluded from adapted CCI) E12, B21.x-B22.x, B24.x, E14, I85.9, I98.x, P29, I79.2, Z49.1, Z49.2, M82

ICD-10 codes used for adaptation of Charlson Comorbidity Index

Supplemental Table 3 Propensity score matching ICD-10 code Characteristic name Before: standardized mean difference Cohort 1 after: % of cohort Cohort 2 after: % of cohort After: standardized mean difference AI Age at index (AI) 0.09303871 100% 100% <0.1 2186-5 Not Hispanic or Latino 0.01386114 83.19% 82.93% <0.1 2106-3 White 0.013364471 78.72% 78.25% <0.1 2054-5 Black or African American 0.001751434 14.08% 14.62% <0.1 UN Unknown ethnicity (UN) 0.00721608 13.89% 14.13% <0.1 2135-2 Hispanic or Latino 0.015690338 2.92% 2.94% <0.1 2028-9 Asian 0.043255243 1.77% 1.70% <0.1 2076-8 Native Hawaiian or Other Pacific Islander 0.000304428 0.38% 0.40% <0.1 1002-5 American Indian or Alaska Native 0.000207706 0.22% 0.21% <0.1 9083 BMI 0.003540847 78.96% 78.63% <0.1 I10 Essential (primary) hypertension 0.08134169 57.21% 56.87% <0.1 E08-E13 Diabetes mellitus 0.10522307 19.84% 19.79% <0.1 E11 Type 2 diabetes mellitus 0.10396519 19.56% 19.53% <0.1 E11.9 Type 2 diabetes mellitus without complications 0.09273047 17.74% 17.75% <0.1 I25.10 Atherosclerotic heart disease of native coronary artery without angina pectoris 0.26752892 14.44% 14.38% <0.1 N18 Chronic kidney disease (CKD) 0.07961401 13.15% 13.11% <0.1 I70-I79 Diseases of arteries, arterioles, and capillaries 0.08853322 12.55% 12.47% <0.1 I60-I69 Cerebrovascular diseases 0.016007513 8.70% 8.54% <0.1 I50 Heart failure 0.091492124 8.55% 8.11% <0.1 J45 Asthma 0.17451213 8.26% 8.48% <0.1 J44 Other chronic obstructive pulmonary disease 0.007154602 8.03% 7.83% <0.1 K70-K77 Diseases of liver 0.046178605 7.51% 7.44% <0.1 E11.6 Type 2 diabetes mellitus with other specified complications 0.07882338 6.82% 6.62% <0.1 I70 Atherosclerosis 0.040540412 4.96% 4.99% <0.1 M81 Osteoporosis without current pathological fracture 0.4244023 4.60% 4.35% <0.1 E11.2 Type 2 diabetes mellitus with kidney complications 0.07015525 4.29% 4.22% <0.1 I73.9 Peripheral vascular disease, unspecified 0.08021869 4.20% 4.02% <0.1 E11.4 Type 2 diabetes mellitus with neurological complications 0.05705401 3.94% 3.91% <0.1 I25.2 Old myocardial infarction 0.12914135 3.90% 3.84% <0.1 I11.0 Hypertensive heart disease with heart failure 0.061771795 3.51% 3.23% <0.1 K76.0 Fatty (change of) liver, not elsewhere classified 0.0236126 3.40% 3.42% <0.1 I21 Acute myocardial infarction 0.08413896 3.35% 3.20% <0.1 Z72.0 Tobacco use 0.06593132 3.20% 3.17% <0.1 M06 Other rheumatoid arthritis 0.17699356 3.19% 3.01% <0.1 J40 Bronchitis, not specified as acute or chronic 0.061653934 3.12% 3.08% <0.1 J43 Emphysema 0.023052668 2.91% 2.90% <0.1 K76.8 Other specified diseases of liver 0.008542072 2.85% 2.87% <0.1 Z95.8 Presence of other cardiac and vascular implants and grafts 0.089386635 2.35% 2.20% <0.1 Z55-Z65 Persons with potential health hazards related to socioeconomic and psychosocial circumstances 0.047443703 2.31% 2.19% <0.1 G45 Transient cerebral ischemic attacks and related syndromes 0.001508629 2.24% 2.25% <0.1 I42.9 Cardiomyopathy, unspecified 0.09544626 2.04% 1.86% <0.1 I42.8 Other cardiomyopathies 0.09315334 1.96% 1.83% <0.1 I71 Aortic aneurysm and dissection 0.1574292 1.84% 1.73% <0.1 I13.0 Hypertensive heart and chronic kidney disease with heart failure and stage 1 through stage 4 chronic kidney disease or unspecified chronic kidney disease 0.04833851 1.75% 1.73% <0.1 E11.8 Type 2 diabetes mellitus with unspecified complications 0.047508586 1.73% 1.69% <0.1 K76.9 Liver disease, unspecified 0.022186145 1.69% 1.69% <0.1 C80.1 Malignant (primary) neoplasm, unspecified 0.007134717 1.62% 1.64% <0.1 K74 Fibrosis and cirrhosis of liver 0.06591101 1.55% 1.53% <0.1 E11.5 Type 2 diabetes mellitus with circulatory complications 0.05308114 1.54% 1.42% <0.1 E11.3 Type 2 diabetes mellitus with ophthalmic complications 0.019891866 1.44% 1.45% <0.1 M80 Osteoporosis with current pathological fracture 0.10710507 1.28% 1.34% <0.1 K25 Gastric ulcer 0.014635305 1.26% 1.22% <0.1 I15 Secondary hypertension 0.0164582 1.25% 1.26% <0.1 I25.5 Ischemic cardiomyopathy 0.117464416 1.14% 1.02% <0.1 I27.8 Other specified pulmonary heart diseases 0.001150849 1.12% 1.01% <0.1 B18 Chronic viral hepatitis 0.080120765 1.11% 1.07% <0.1 J60-J70 Lung diseases due to external agents 0.029337624 1.03% 1.02% <0.1 I42.5 Other restrictive cardiomyopathy 0.06866931 1.03% 0.98% <0.1 E10.4 Type 1 diabetes mellitus with neurological complications 0.030510664 0.95% 0.94% <0.1 E13.4 Other specified diabetes mellitus with neurological complications 0.03325113 0.92% 0.91% <0.1 I12.0 Hypertensive chronic kidney disease with stage 5 chronic kidney disease or end stage renal disease 0.044459436 0.82% 0.84% <0.1 M35.3 Polymyalgia rheumatica 0.036451396 0.80% 0.79% <0.1 K27 Peptic ulcer, site unspecified 0.006956438 0.76% 0.76% <0.1 J47 Bronchiectasis 0.037709255 0.76% 0.69% <0.1 I77.1 Stricture of artery 0.016310673 0.74% 0.69% <0.1 E10.9 Type 1 diabetes mellitus without complications 0.02156937 0.73% 0.65% <0.1 M05 Rheumatoid arthritis with rheumatoid factor 0.10806804 0.73% 0.70% <0.1 N19 Unspecified kidney failure 0.04624667 0.69% 0.70% <0.1 K70.3 Alcoholic cirrhosis of liver 0.079953395 0.66% 0.61% <0.1 J42 Unspecified chronic bronchitis 0.030202676 0.66% 0.64% <0.1 K76.6 Portal hypertension 0.055476364 0.64% 0.65% <0.1 K72 Hepatic failure, not elsewhere classified 0.04292999 0.63% 0.64% <0.1 G81 Hemiplegia and hemiparesis 0.006484089 0.61% 0.57% <0.1 Z99.2 Dependence on renal dialysis 0.033356454 0.61% 0.66% <0.1 I42.0 Dilated cardiomyopathy 0.05840778 0.60% 0.56% <0.1 K26 Duodenal ulcer 0.06509399 0.57% 0.60% <0.1 Z94.0 Kidney transplant status 0.05038971 0.54% 0.55% <0.1 J41 Simple and mucopurulent chronic bronchitis 0.00367931 0.52% 0.54% <0.1 E13.9 Other specified diabetes mellitus without complications 0.01683673 0.50% 0.49% <0.1 B20-B20 Human immunodeficiency virus (HIV) disease (B20) 0.09872614 0.48% 0.44% <0.1 I13.1 Hypertensive heart and chronic kidney disease without heart failure 0.027477937 0.45% 0.47% <0.1 E10.6 Type 1 diabetes mellitus with other specified complications 0.026096078 0.45% 0.40% <0.1 K72.9 Hepatic failure, unspecified 0.04491207 0.41% 0.42% <0.1 I85.0 Esophageal varices 0.04319256 0.36% 0.39% <0.1 N25.0 Renal osteodystrophy 0.017381856 0.35% 0.34% <0.1 M32 Systemic lupus erythematosus (SLE) 0.15744372 0.35% 0.32% <0.1 E11.0 Type 2 diabetes mellitus with hyperosmolarity 0.02392833 0.33% 0.31% <0.1 R54 Age-related physical debility 0.030876666 0.29% 0.33% <0.1 E63 Other nutritional deficiencies 0.01147259 0.29% 0.29% <0.1 K70.1 Alcoholic hepatitis 0.03860219 0.27% 0.25% <0.1 Z94.4 Liver transplant status 0.052426532 0.27% 0.29% <0.1 I43 Cardiomyopathy in diseases classified elsewhere 0.027464632 0.26% 0.26% <0.1 I73.8 Other specified peripheral vascular diseases 0.018975774 0.26% 0.28% <0.1 I13.2 Hypertensive heart and chronic kidney disease with heart failure and with stage 5 chronic kidney disease or end stage renal disease 0.028738411 0.25% 0.26% <0.1 E63.9 Nutritional deficiency, unspecified 0.00688869 0.23% 0.24% <0.1 G82 Paraplegia (paraparesis) and quadriplegia (quadriparesis) 0.022867465 0.22% 0.22% <0.1 K55.9 Vascular disorder of intestine, unspecified 0.018867034 0.21% 0.21% <0.1 E10.2 Type 1 diabetes mellitus with kidney complications 0.017680315 0.21% 0.18% <0.1 K70.9 Alcoholic liver disease, unspecified 0.033337377 0.19% 0.17% <0.1 K72.1 Chronic hepatic failure 0.034303084 0.19% 0.20% <0.1 E13.6 Other specified diabetes mellitus with other specified complications 0.020468568 0.19% 0.21% <0.1 K28 Gastrojejunal ulcer 0.003309273 0.18% 0.16% <0.1 E11.1 Type 2 diabetes mellitus with ketoacidosis 0.01218536 0.16% 0.18% <0.1 E10.3 Type 1 diabetes mellitus with ophthalmic complications 0.022044873 0.16% 0.14% <0.1 Z95.9 Presence of cardiac and vascular implant and graft, unspecified 0.012073873 0.14% 0.14% <0.1 G83.4 Cauda equina syndrome 0.012899983 0.14% 0.18% <0.1 K73 Chronic hepatitis, not elsewhere classified 0.016151981 0.14% 0.12% <0.1 K55.1 Chronic vascular disorders of intestine 0.010709321 0.14% 0.13% <0.1 I42.7 Cardiomyopathy due to drug and external agent 0.000498729 0.12% 0.11% <0.1 E13.1 Other specified diabetes mellitus with ketoacidosis 0.009991395 0.12% 0.13% <0.1 E10.8 Type 1 diabetes mellitus with unspecified complications 4.12186E-06 0.11% 0.11% <0.1 M33 Dermatopolymyositis 0.013817623 0.11% 0.12% <0.1 M34 Systemic sclerosis [scleroderma] 0.05081338 0.11% 0.14% <0.1 I27.9 Pulmonary heart disease, unspecified 0.005778024 0.11% 0.11% <0.1 E10.5 Type 1 diabetes mellitus with circulatory complications 0.018261774 0.10% 0.08% <0.1 E10.1 Type 1 diabetes mellitus with ketoacidosis 0.009557457 0.08% 0.09% <0.1 Z49.0 Preparatory care for renal dialysis 0.003154584 0.08% 0.09% <0.1 K70.0 Alcoholic fatty liver 0.033738267 0.07% 0.06% <0.1 G83.9 Paralytic syndrome, unspecified 0.011464326 0.07% 0.05% <0.1 I86.4 Gastric varices 0.01937838 0.07% 0.06% <0.1 K76.7 Hepatorenal syndrome 0.02139664 0.07% 0.07% <0.1 G83.1 Monoplegia of lower limb 0.014896353 0.07% 0.05% <0.1 N03.3 Chronic nephritic syndrome with diffuse mesangial proliferative glomerulonephritis 0.006567977 0.07% 0.05% <0.1 E13.8 Other specified diabetes mellitus with unspecified complications 0.008838696 0.06% 0.07% <0.1 K76.2 Central hemorrhagic necrosis of liver 0.003881067 0.06% 0.08% <0.1 G83.2 Monoplegia of upper limb 0.009757021 0.06% 0.06% <0.1 K70.4 Alcoholic hepatic failure 0.028182138 0.05% 0.05% <0.1 E13.2 Other specified diabetes mellitus with kidney complications 0.003916801 0.05% 0.06% <0.1 I22 Subsequent ST-segment elevation (STEMI) and non-ST-segment elevation (NSTEMI) myocardial infarction 0.011183402 0.05% 0.04% <0.1 J67 Hypersensitivity pneumonitis due to organic dust 0.009636628 0.05% 0.05% <0.1 E13.3 Other specified diabetes mellitus with ophthalmic complications 0.009757021 0.05% 0.06% <0.1 H34.0 Transient retinal artery occlusion 0.001207267 0.05% 0.05% <0.1 G46 Vascular syndromes of brain in cerebrovascular diseases 0.006143574 0.04% 0.04% <0.1 I09.9 Rheumatic heart disease, unspecified 0.009838371 0.04% 0.04% <0.1 I42.6 Alcoholic cardiomyopathy 0.026666112 0.04% 0.04% <0.1 K70.2 Alcoholic fibrosis and sclerosis of liver 0.005191918 0.04% 0.04% <0.1 K71.5 Toxic liver disease with chronic active hepatitis 0.005191918 0.04% 0.04% <0.1 K71.7 Toxic liver disease with fibrosis and cirrhosis of liver 0.005191918 0.04% 0.04% <0.1 K71.1 Toxic liver disease with hepatic necrosis 0.005191918 0.04% 0.04% <0.1 K76.5 Hepatic veno-occlusive disease 0.005191918 0.04% 0.04% <0.1 K55.8 Other vascular disorders of intestine 0.001994597 0.04% 0.04% <0.1 G11.4 Hereditary spastic paraplegia 0.005191918 0.04% 0.04% <0.1 G80.1 Spastic diplegic cerebral palsy 0.005191918 0.04% 0.04% <0.1 G80.2 Spastic hemiplegic cerebral palsy 0.005191918 0.04% 0.04% <0.1 G83.3 Monoplegia, unspecified 0.005191918 0.04% 0% <0.1 N03.2 Chronic nephritic syndrome with diffuse membranous glomerulonephritis 0.005191918 0.04% 0.04% <0.1 N05.2 Unspecified nephritic syndrome with diffuse membranous glomerulonephritis 0.015125674 0.04% 0.04% <0.1 N05.5 Unspecified nephritic syndrome with diffuse mesangiocapillary glomerulonephritis 0.005191918 0.04% 0.04% <0.1 N05.7 Unspecified nephritic syndrome with diffuse crescentic glomerulonephritis 0.005191918 0.04% 0.04% <0.1 M31.5 Giant cell arteritis with polymyalgia rheumatica 0.005191918 0.04% 0.04% <0.1 M35.1 Other overlap syndromes 0.05861349 0.04% 0.04% <0.1 J60 Coal worker's pneumoconiosis 0.010551658 0.04% 0.04% <0.1 J61 Pneumoconiosis due to asbestos and other mineral fibers 0.040973343 0.04% 0.04% <0.1 J62 Pneumoconiosis due to dust containing silica 0.005191918 0.04% 0% <0.1 J63 Pneumoconiosis due to other inorganic dusts 0.005191918 0.04% 0.04% <0.1 J66 Airway disease due to specific organic dust 0.027462788 0.04% 0% <0.1 J70.1 Chronic and other pulmonary manifestations due to radiation 0.008682114 0.04% 0.04% <0.1 E13.0 Other specified diabetes mellitus with hyperosmolarity 0.021716576 0.04% 0.04% <0.1 E13.5 Other specified diabetes mellitus with circulatory complications 0.005191918 0.04% 0.04% <0.1 I73.1 Thromboangiitis obliterans [Buerger's disease] 0.005191918 0.04% 0.04% <0.1 K71.3 Toxic liver disease with chronic persistent hepatitis 0.027462788 0% 0% K71.4 Toxic liver disease with chronic lobular hepatitis 0% 0% K76.3 Infarction of liver 0.027462788 0% 0% K76.4 Peliosis hepatis 0% 0% G04.1 Tropical spastic paraplegia 0% 0% G83.0 Diplegia of upper limbs 0.005191918 0% 0.04% <0.1 N03.5 Chronic nephritic syndrome with diffuse mesangiocapillary glomerulonephritis 0.005191918 0% 0.04% <0.1 N03.7 Chronic nephritic syndrome with diffuse crescentic glomerulonephritis 0% 0% N03.4 Chronic nephritic syndrome with diffuse endocapillary proliferative glomerulonephritis 0% 0% N03.6 Chronic nephritic syndrome with dense deposit disease 0% 0% N05.3 Unspecified nephritic syndrome with diffuse mesangial proliferative glomerulonephritis 0.023780894 0% 0% N05.6 Unspecified nephritic syndrome with dense deposit disease 0% 0% N05.4 Unspecified nephritic syndrome with diffuse endocapillary proliferative glomerulonephritis 0% 0% M36.0 Dermato(poly)myositis in neoplastic disease 0% 0% J64 Unspecified pneumoconiosis 0.027462788 0% 0% J65 Pneumoconiosis associated with tuberculosis 0% 0% J68.4 Chronic respiratory conditions due to chemicals, gases, fumes, and vapors 0.023780894 0% 0.04% <0.1 J70.3 Chronic drug-induced interstitial lung disorders 0.005191918 0% 0.04% <0.1 I79.0 Aneurysm of aorta in diseases classified elsewhere 0% 0% BMI, body mass index.

Propensity score matching

BMI, body mass index.

References

Associated Data

Supplementary Materials

Conflict of Interest Statement for Layton

Conflict of Interest Statement for Rehman

Conflict of Interest Statement for Prasad

Conflict of Interest Statement for Patel