Abstract
Introduction
The purpose of this study is to investigate the relationship between preoperative hyponatremia and postoperative complications in aseptic revision hip and knee arthroplasty.
Methods
Aseptic revision arthroplasties from 2007 to 2016 were collected using a large, United States database. Patients were stratified into eunatremic and hyponatremic cohorts. Rates of complications were collected and analyzed.
Results
25,517 surgeries were evaluated. Preoperative hyponatremia was independently associated with organ/space infections (OR= 2.316; p<0.001), postoperative blood transfusions (OR= 1.467; p<0.001), pneumonia (OR= 2.05; p =0.002), sepsis (OR= 2.533; p<0.001), extended length-of-stay (OR= 2.221; p<0.001), minor complications (OR= 1.549; p<0.001), wound complications (OR= 1.505; p=0.001), pulmonary complications (OR= 1.72; p=0.007), and sepsis complications (OR= 2.305; p<0.001).
Conclusion
Hyponatremia is an independent risk factor for several postoperative complications in aseptic revision hip and knee arthroplasty.
Keywords: Hyponatremia, Revision knee arthroplasty, Revision hip arthroplasty, Hip, Knee, Postoperative complications
1. Introduction
Hyponatremia is defined as a serum sodium (Na) < 135 mEq/L and is the most common electrolyte disturbance seen in the hospital setting.1,2 Both preoperative and postoperative hyponatremia in orthopedic patients have been linked to longer and costlier hospitalizations.3 Patients who present with preoperative hyponatremia before total knee arthroplasty (TKA) have longer hospital stays and are more likely to require revision arthroplasty.4 Certain medications like diuretics, proton pump inhibitors (PPIs), and selective serotonin reuptake inhibitors (SSRIs) increase the risk of postoperative hyponatremia, and hip surgery itself has also been proposed as a risk factor for developing new-onset postoperative hyponatremia.5 Thus, both preoperative and postoperative hyponatremia are associated with adverse postoperative complications in the orthopedic population.
To date, no studies have investigated the association between preoperative hyponatremia and postoperative complications in aseptic revision arthroplasties. Compared to primary arthroplasties, revision arthroplasties are more complex procedures that require removal of the primary prostheses, reconstruction of the surrounding bone and soft tissue, and implantation of specialized prostheses, often with additional support.6 Revision arthroplasties are therefore more technically demanding than primary arthroplasties and are associated with a higher rate of complications.7 Over the next decade in the United States, the number of revision total hip arthroplasties is projected to increase by 43% and the number of revision total knee arthroplasties is projected to increase by 121%.8,9 Given that the number of both septic and aseptic revision arthroplasties is projected to increase significantly, careful analysis of modifiable risk factors for postoperative complications is necessary to improve outcomes for future patients. The purpose of this study was to investigate the relationship between preoperative hyponatremia and postoperative complications in patients undergoing aseptic revision hip and knee arthroplasty.
2. Methods
Data was collected using the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database for all revision hip and knee arthroplasties between 2007 and 2016. The American College of Surgeons NSQIP database is a national database with over 600 participating hospitals that focuses on risk-adjusted outcomes and measures to improve the quality of surgical care. Current Procedural Terminology (CPT) codes were used to identify appropriate patients who underwent a revision total knee arthroplasty or revision total hip arthroplasty (THA). CPT codes 27486 and 27487 were used to identify patients that underwent a revision TKA, while CPT codes 27134, 27137 and 27138 were used to identify those that underwent revision THA. Patients were excluded if they underwent revision TKA or THA for periprosthetic joint infection, revision hemiarthroplasty of the hip, or conversion hip arthroplasty. Furthermore, patients were also excluded if there was missing information or variables, or if patients were less than 18 years of age.
Eunatremia was defined as Na level between 135 and 145 mEq/L and hyponatremia was defined as Na <135 mEq/L. Patients were then stratified as either hyponatremic or eunatremic prior to revision joint replacement. Demographics, comorbidities, and postoperative complication rates were collected. Demographics included sex, age, race, and other previously identified known risk factors for surgical complications. These risk factors included smoking status, diabetic status, body mass index (BMI), dyspnea status, American Society of Anesthesiologists (ASA) class, type of anesthesia used, and preoperative functional status. Medical comorbidity data collected included congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), hypertension (HTN), renal failure, dialysis dependence, steroid use, weight loss, bleeding disorders, and preoperative transfusions.
Thirty-day postoperative complications included surgical site wound infection, deep wound infection, organ or space infections, wound dehiscence, pneumonia, reintubation, pulmonary embolism, failure to wean off ventilator for greater than 48 h, renal failure, renal insufficiency, urinary tract infection (UTI), stroke, cardiac arrest, myocardial infarction, deep vein thrombosis, sepsis and septic shock, and all-cause death. Complications were also added for each cohort and labeled as all-cause postoperative complications. The analyzed postoperative outcomes within 30 days included extended length of stay and unplanned return to the operating room. Extended length of stay was determined to be an inpatient hospital stay >7 days.
Statistical Package for the Social Sciences (SPSS; version 22; Armonk, NY) software was used to conduct univariate and multivariate analyses on demographic data, comorbidities, and postoperative complications. Chi-square test and one-way analysis of variance (ANOVA) were used to perform univariate testing where appropriate. Risk factors with P < 0.2 in univariate analyses were selected for multivariate analyses to determine independent risk factors for postoperative complications, return to the operating room, and extended length of stay. Multivariate analysis was performed using binary logistic regression analysis on postoperative complications, unplanned return to OR, and extended length of stay within 30 days using forward stepwise. The results of the multivariate analyses were reported with P values and odds ratios with 95% confidence intervals. A p value of <0.05 was deemed the cutoff value for statistical significance.
3. Results
3.1. Demographics
From 2007 to 2016, 25,517 patients underwent revision THA or revision TKA whose records were analyzed in the NSQIP database. In total, 23,822 patients were eunatremic and 1695 were hyponatremic preoperatively. After stratifying by Na status, hyponatremic patients are more likely to be white (p < 0.001), Type 1 Diabetics (p < 0.001), smokers (p < 0.001), over 70 (p < 0.001), and having BMI from normal to overweight range (p < 0.001). These patients were also more likely to have received general anesthesia (p = 0.001), to be partially dependent (p < 0.001), and to have ASA ratings 3 or 4 (p < 0.001). Demographics and clinical characteristics of these patients are shown in Table 1.
Table 1.
Demographic and Clinical Characteristics stratified by Na status.
| Demographics and Clinical Characteristics | |||||
|---|---|---|---|---|---|
| Demographics | Hyponatremic |
Eunatremic |
P-value |
||
| 1695 |
23822 |
||||
| N | % | N | % | ||
| Sex | 0.655 | ||||
| Female | 998 | 58.90% | 13759 | 57.80% | |
| Male | 696 | 41.10% | 10051 | 42.20% | |
| Race | <0.001 | ||||
| White | 1306 | 77.10% | 17869 | 75.00% | |
| Black or African American | 93 | 5.50% | 2391 | 10.00% | |
| Hispanic | 48 | 2.80% | 897 | 3.80% | |
| American Indian or Alaska Native | 15 | 0.90% | 117 | 0.50% | |
| Asian | 19 | 1.10% | 229 | 1.00% | |
| Native Hawaiian or Pacific Islander | 9 | 0.50% | 60 | 0.30% | |
| Anesthesia | 0.001 | ||||
| General | 1191 | 70.30% | 15539 | 65.20% | |
| Regional | 415 | 24.50% | 6737 | 28.30% | |
| MAC/IV Sedation | 78 | 4.60% | 1509 | 6.30% | |
| Diabetes Mellitus Status | <0.001 | ||||
| No DM | 1351 | 79.70% | 19708 | 82.70% | |
| NIDDM | 194 | 11.40% | 2833 | 11.90% | |
| IDDM | 150 | 8.80% | 1281 | 5.40% | |
| ASA Score | <0.001 | ||||
| 1 or 2 | 1536 | 90.60% | 22499 | 94.40% | |
| 3 or 4 | 147 | 8.70% | 929 | 3.90% | |
| Smoking | 302 | 17.80% | 3134 | 13.20% | <0.001 |
| Dyspnea | 0.029 | ||||
| No dyspnea | 1569 | 92.60% | 22187 | 93.10% | |
| Moderate exertion | 113 | 6.70% | 1550 | 6.50% | |
| At rest | 13 | 0.80% | 85 | 0.40% | |
| Functional Status preoperatively | <0.001 | ||||
| Independent | 1526 | 90.00% | 22500 | 94.50% | |
| Partially Dependent | 136 | 8.00% | 1077 | 4.50% | |
| Totally Dependent | 20 | 1.20% | 91 | 0.40% | |
| Age | <0.001 | ||||
| <50 | 94 | 5.50% | 1903 | 8.00% | |
| 51–60 | 296 | 17.50% | 5158 | 21.70% | |
| 61–70 | 435 | 25.70% | 7691 | 32.30% | |
| 71–80 | 481 | 28.40% | 5984 | 25.10% | |
| 81+ | 352 | 20.80% | 2920 | 12.30% | |
| Body Mass Index | <0.001 | ||||
| 18.5–30 | 921 | 54.30% | 10509 | 44.10% | |
| 30–34.9 | 351 | 20.70% | 6177 | 25.90% | |
| 35–39.9 | 194 | 11.40% | 3723 | 15.60% | |
| >40 | 153 | 9.00% | 2978 | 12.50% | |
*Bolding indicates significance.
3.2. Comorbidities
Patients that presented with hyponatremia preoperatively were more likely to have COPD (p < 0.001), CHF (p < 0.001), hypertension (p < 0.001), renal failure (p = 0.017), dialysis (p < 0.001), steroid use (p = 0.043), bleeding disorders (p < 0.001), and preoperative blood transfusions (p < 0.001). Comorbidities of these patients are shown in Table 2.
Table 2.
Comorbidities stratified by Na status.
| Comorbidities | Hyponatremic |
Eunatremic |
P-value |
||
|---|---|---|---|---|---|
| 1695 |
23822 |
||||
| N | % | N | % | ||
| COPD | 153 | 9.00% | 1296 | 5.40% | <0.001 |
| CHF | 22 | 1.30% | 137 | 0.60% | <0.001 |
| Hypertension | 1288 | 76.00% | 15047 | 63.20% | <0.001 |
| Renal Failure | 3 | 0.20% | 10 | 0.00% | 0.017 |
| Dialysis | 24 | 1.40% | 94 | 0.40% | <0.001 |
| Steroid use | 107 | 6.30% | 1234 | 5.20% | 0.043 |
| Weight loss | 10 | 0.60% | 103 | 0.40% | 0.345 |
| Bleeding disorder | 130 | 7.70% | 1063 | 4.50% | <0.001 |
| Transfusion | 48 | 2.80% | 200 | 0.80% | <0.001 |
| OR Time >3 Hours | 386 | 22.80% | 5136 | 21.60% | 0.242 |
*Bolding indicates significance.
3.3. Complications
Hyponatremic patients were found to be more likely to have any complication (p < 0.001), deep wound infections (p < 0.001), organ/space infections (p < 0.001), pneumonia (p < 0.001), UTIs (p < 0.001), myocardial infarctions (p = 0.045), sepsis (p < 0.001), death (p < 0.001), extended length-of-stay (p < 0.001), major complications (p < 0.001), minor complications (p < 0.001), wound complications (p < 0.001), pulmonary complications (p < 0.001), and sepsis complications (p < 0.001). Complications of these patients are shown in Table 3.
Table 3.
Univariate Analysis of Postoperative Complications stratified by Na status.
| Complications | Hyponatremic |
Eunatremic |
P-value |
||
|---|---|---|---|---|---|
| 1695 |
23822 |
||||
| N | % | N | % | ||
| Any complication | 271 | 16.00% | 2185 | 9.20% | <0.001 |
| Superficial surgical site wound | 16 | 0.90% | 231 | 1.00% | 0.917 |
| Deep Wound Infection | 19 | 1.10% | 231 | 1.00% | <0.001 |
| Organ/Space Infection | 48 | 2.80% | 293 | 1.20% | <0.001 |
| Wound Dehiscence | 11 | 0.60% | 93 | 0.40% | 0.106 |
| Pneumonia | 28 | 1.70% | 127 | 0.50% | <0.001 |
| Reintubation | 10 | 0.60% | 65 | 0.30% | 0.02 |
| Pulmonary Embolism | 6 | 0.40% | 96 | 0.40% | 0.757 |
| Failure to wean off Ventilator >48 h | 1 | 0.20% | 8 | 0.20% | 0.075 |
| Renal insufficiency | 3 | 0.20% | 58 | 0.20% | 0.588 |
| Renal failure | 3 | 0.20% | 24 | 0.10% | 0.351 |
| Urinary tract Infection | 35 | 2.10% | 271 | 1.10% | <0.001 |
| Stroke | 1 | 0.10% | 37 | 0.20% | 0.32 |
| Cardiac Arrest | 4 | 0.20% | 41 | 0.20% | 0.545 |
| Myocardial Infarction | 12 | 0.70% | 92 | 0.40% | 0.045 |
| Deep Vein Thrombosis | 10 | 0.60% | 184 | 0.80% | 0.403 |
| Sepsis | 41 | 2.40% | 206 | 0.90% | <0.001 |
| Septic Shock | 5 | 0.30% | 41 | 0.20% | 0.249 |
| Death | 18 | 1.10% | 87 | 0.40% | <0.001 |
| Other Outcomes | |||||
| Extended LOS (>7 days) | 307 | 18.10% | 1584 | 6.60% | <0.001 |
| Return to Operating Room | 14 | 0.80% | 332 | 1.40% | 0.799 |
| Readmission | 11 | 0.60% | 126 | 0.50% | 0.829 |
| Reoperation | 0 | 0.00% | 9 | 0.00% | 0.502 |
| Major Complication | 78 | 4.60% | 643 | 2.70% | <0.001 |
| Minor Complication | 573 | 33.80% | 5024 | 21.10% | <0.001 |
| Wound Complication | 87 | 5.10% | 794 | 3.30% | <0.001 |
| Pulmonary complications | 36 | 2.10% | 186 | 0.80% | <0.001 |
| Cardiac Complications | 15 | 0.90% | 126 | 0.50% | 0.056 |
| Renal Complications | 6 | 0.40% | 80 | 0.30% | 0.901 |
| Thromboembolic complications | 16 | 0.90% | 296 | 1.20% | 0.28 |
| Sepsis Complications | 46 | 2.70% | 241 | 1.00% | <0.001 |
*Bolding indicates significance.
3.4. Multivariate analysis
Following multivariate analysis, hyponatremia was found to be an independent risk factor for organ/space infections (p < 0.001; OR = 2.316; 95% CI [1.681, 3.192]), postoperative blood transfusions (p < 0.001; OR = 1.467; 95% CI [1.302, 1.652]), pneumonia (p = 0.002; OR = 2.05; 95% CI [1.314, 3.199]), sepsis (p < 0.001; OR = 2.533; 95% CI [1.776, 3.614]), extended length-of-stay (p < 0.001; OR = 2.221; 95% CI [1.909, 2.584]), minor complications (p < 0.001; OR = 1.549; 95% CI [1.384, 1.733]), wound complications (p = 0.001; OR = 1.505; 95% CI [1.19, 1.904]), pulmonary complications (p = 0.007; OR = 1.72; 95% CI [1.161, 2.547]), and sepsis complications (p < 0.001; OR = 2.305; 95% CI [1.648, 3.223]). Results of the multivariate analysis are shown in Table 4.
Table 4.
Multivariate Analysis of Relationship between Na status and Postoperative Complications.
| Postoperative Complication | p-value | Odds Ratios | 95% CI |
|
|---|---|---|---|---|
| low | high | |||
| Deep Wound Infection | 0.713 | 1.094 | 0.678 | 1.766 |
| Organ/Space Infection | <0.001 | 2.316 | 1.681 | 3.192 |
| Postoperative Blood Transfusion | <0.001 | 1.467 | 1.302 | 1.652 |
| Pneumonia | 0.002 | 2.050 | 1.314 | 3.199 |
| Urinary tract Infection | 0.065 | 1.432 | 0.978 | 2.096 |
| Myocardial Infarction | 0.584 | 1.197 | 0.629 | 2.279 |
| Sepsis | <0.001 | 2.533 | 1.776 | 3.614 |
| Death | 0.339 | 1.330 | 0.741 | 2.385 |
| Extended LOS (>7 days) | <0.001 | 2.221 | 1.909 | 2.584 |
| Major Complication | 0.066 | 1.273 | 0.984 | 1.646 |
| Minor Complication | <0.001 | 1.549 | 1.384 | 1.733 |
| Wound Complication | 0.001 | 1.505 | 1.190 | 1.904 |
| Pulmonary complications | 0.007 | 1.720 | 1.161 | 2.547 |
| Sepsis Complications | <0.001 | 2.305 | 1.648 | 3.223 |
*Bolding indicates significance.
4. Discussion
Previous studies have established the connection between hyponatremia and postoperative complications in orthopedic patients. For instance, in patients undergoing primary total knee arthroplasty, preoperative hyponatremia has been associated with extended hospital stays and likelihood of revision.4 These patients also have greater 30-day mortality (hazard ratio [HR] = 2.47 for severe hyponatremia; HR = 1.80 for mild hyponatremia).10
While such studies established the connection between hyponatremia and adverse outcomes in orthopedic patients undergoing primary joint arthroplasty, our present analysis establishes hyponatremia as an independent risk factor for complications in patients undergoing revision hip and knee arthroplasty. We have found that preoperative hyponatremia is correlated with postoperative infections, pneumonia, UTIs, sepsis, extended length-of-stay, wound complications, pulmonary complications, sepsis complications, and death. Because hyponatremia is an independent risk factor, we propose that treating the hyponatremia alone could potentially reduce the incidence of these complications.
Most aseptic revisions are elective procedures. In the era of value-based care, the identification of modifiable risk factors has become important in minimizing postoperative complications. In this study, we demonstrate that preoperative hyponatremia, which is a modifiable risk factor in most cases, directly correlates with multiple postoperative complications. As such, correction or optimization of this abnormality before elective revision surgery should be considered when appropriate.
While this is the first study to examine the effects of preoperative hyponatremia on postoperative complication rates in the revision arthroplasty population, hyponatremia has been shown to be a significant risk factor for numerous postoperative complications after major surgery. In a NSQIP database study examining nearly 1 million patients, preoperative hyponatremia was associated with a four-fold increase in the risk of 30-day mortality following major surgery, and this effect was even more significant when looking specifically at patients undergoing elective surgery.11 Similar to our study, preoperative hyponatremia was found to be an independent negative prognostic risk factor for elective surgery, and it was also associated with an increased risk of major coronary events, wound infections, pneumonia, and increased hospital length-of-stay.
There are several limitations to our study of hyponatremia in patients undergoing revision knee and hip arthroplasties. Because outcomes are only recorded within a 30-day period in the NSQIP database, we were unable to capture adverse outcomes outside of this time period. In addition, differences in postoperative complications between the first arthroplasty and the revision arthroplasty for each patient were not studied and should be addressed in future studies. Common indications for revision arthroplasties for these patients and stratification of postoperative complications based on surgical approach and implant type could also be investigated. Finally, due to the data available for analysis, this study is unable to determine if correction of hyponatremia would allow the complication rate to more closely mirror that of the general population undergoing revision arthroplasty. Further research is needed to determine if correction of the electrolyte abnormality alone is sufficient for reducing the postoperative complication rate, and if it is, what the optimal timeline and rate of correction should be.
In conclusion, hyponatremia is an independent risk factor for postoperative complications including organ/space infections, postoperative blood transfusions, pneumonia, extended length-of-stay, minor complications, wound complications, pulmonary complications, and sepsis complications in patients undergoing revision hip and knee arthroplasty. Patients who present with preoperative hyponatremia should be medically optimized preoperatively and closely surveilled postoperatively in order to minimize the risk of postoperative complications.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of competing interest
The authors of this manuscript certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) directly related to the subject matter or materials discussed in this manuscript.
Acknowledgements
None.
References
- 1.Upadhyay A., Jaber B.L., Madias N.E. Incidence and prevalence of hyponatremia. Am J Med. 2006;119:S30–S35. doi: 10.1016/j.amjmed.2006.05.005. [DOI] [PubMed] [Google Scholar]
- 2.Thompson C., Berl T., Tejedor A., Johannsson G. Differential diagnosis of hyponatremia. Best Pract Res Clin Endocrinol Metabol. 2012;26:S7–S15. doi: 10.1016/S1521-690X(12)70003-9. [DOI] [PubMed] [Google Scholar]
- 3.Hennrikus E., Ou G., Kinney B. Prevalence, timing, causes, and outcomes of hyponatremia in hospitalized orthopaedic surgery patients. J Bone Joint Surg Am. 2015;97:1824–1832. doi: 10.2106/JBJS.O.00103. [DOI] [PubMed] [Google Scholar]
- 4.Abola M.V., Tanenbaum J.E., Bomberger T.T., Knapik D.M., Fitzgerald S.J., Wera G.D. Preoperative hyponatremia is associated with reoperation and prolonged length of hospital stay following total knee arthroplasty. J Knee Surg. 2019;32:344–351. doi: 10.1055/s-0038-1641156. [DOI] [PubMed] [Google Scholar]
- 5.Rudge J.E., Kim D. New onset hyponatraemia after surgery for traumatic hip fracture. Age Ageing. 2014;43:821–826. doi: 10.1093/ageing/afu069. [DOI] [PubMed] [Google Scholar]
- 6.Fehring T.K., Griffin W.L. Revision of failed cementless total knee implants with cement. Clin Orthop Relat Res. 1998;356:34–38. doi: 10.1097/00003086-199811000-00007. [DOI] [PubMed] [Google Scholar]
- 7.Boddapati V., Fu M.C., Mayman D.J., Su E.P., Sculco P.K., McLawhorn A.S. Revision total knee arthroplasty for periprosthetic joint infection is associated with increased postoperative morbidity and mortality relative to noninfectious revisions. J Arthroplasty. 2018;33:521–526. doi: 10.1016/j.arth.2017.09.021. [DOI] [PubMed] [Google Scholar]
- 8.Kurtz S., Ong K., Lau E., Mowat F., Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89:780–785. doi: 10.2106/JBJS.F.00222. [DOI] [PubMed] [Google Scholar]
- 9.Hamilton D.F., Simpson P.M., Patton J.T., Howie C.R., Burnett R. Aseptic revision knee arthroplasty with total stabilizer prostheses achieves similar functional outcomes to primary total knee arthroplasty at 2 Years: a longitudinal cohort study. J Arthroplasty. 2017;32:1234–1240. doi: 10.1016/j.arth.2016.10.028. [DOI] [PubMed] [Google Scholar]
- 10.McCausland F.R., Wright J., Waikar S.S. Association of serum sodium with morbidity and mortality in hospitalized patients undergoing major orthopedic surgery. J Hosp Med. 2014;9:297–302. doi: 10.1002/jhm.2168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Leung A.A., McAlister F.A., Rogers S.O., Pazo V., Wright A., Bates D.W. Preoperative hyponatremia and perioperative complications. Arch Intern Med. 2012;172:1474–1481. doi: 10.1001/archinternmed.2012.3992. [DOI] [PubMed] [Google Scholar]