Chronic kidney disease (CKD) affects 5% of the population, including patients dependent on dialysis.1 CKD, defined as estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2, is predictive of increased all-cause2–8 and cardiovascular mortality in individuals with vascular disease2,3 and asymptomatic middle-aged individuals.4,5 CKD integrates several key pathophysiologic etiologies that may contribute to postoperative morbidity/mortality,6 including increased levels of inflammatory factors,7 elevated plasma homocysteine,8 enhanced coagulability,9,10 excess arterial calcification,11 and endothelial dysfunction.12 Meta-analysis of elective, noncardiac (largely vascular) surgical studies shows that CKD is an independent risk factor for postoperative death and cardiovascular events.12 However, the relationship of chronic, mild renal dysfunction with postoperative morbidity is unclear, particularly in apparently lower-risk patients undergoing moderate-risk surgery. Because the detrimental impact of postoperative morbidity on longer-term outcomes (including mortality) is now well established,13 assessing the relationship of impaired renal function with postoperative morbidity may be particularly instructive. In the cardiac surgical setting, the eGFR is of value in predicting mortality in patients undergoing cardiac surgery.14,15 Major joint replacement surgery is also associated with perioperative morbidity, particularly in an increasingly elderly population with significant comorbidities.16–18 We hypothesized that CKD would be associated with excess postoperative morbidity and prolonged hospital stay after elective, moderate-risk orthopedic procedures.
Supplementary Table 1 (see Supplemental Digital Content 1, http://links.lww.com/AA/A164) details compliance with the STROBE checklist19 of items recommended for the reporting of observational studies.
In accordance with local Research Ethical Committee approval and the Ethical Principles for Medical Research Involving Human Subjects as outlined in the Declaration of Helsinki, written informed consent was obtained preoperatively from patients presenting for elective major joint (knee and hip) replacement, including revision procedures, aged 50 years and older, between March 2004 and April 2005 at University College London Hospital, United Kingdom.
Surgery and anesthesia were conducted by attending staff. Anesthesia, standardized antibiotic prophylaxis, fluid therapy and physiotherapy care were delivered according to usual local standard of care. Epidural anesthesia/analgesia was discontinued within 36 hours postoperatively to enable mobilization and physiotherapy. Morphine patient-controlled analgesia was administered to patients who had undergone procedures under spinal anesthesia. An acute pain service was available for consultation. Oral fluid and solid intake was encouraged to be resumed on postoperative day 1.
Definition of Renal Function and Perioperative Risk Factors
Preoperative creatinine levels were obtained from routine preoperative blood tests. Glomerular filtration rate (GFR) was estimated from the Modification of Diet in Renal Disease equation20,21:
The product of this equation was multiplied by a correction factor of 0.742 for women and 1.212 for African American subjects. CKD was defined as eGFR <60 mL/min per 1.73 m2, according to the Kidney Disease Outcomes Quality Initiative classification.22 CKD patients were further categorized a priori into 3 subgroups depending on their eGFR: 59 to 50, 49 to 40, or 39 to 20 mL/min/1.73 m2, according to recent population-based studies exploring long-term cardiovascular risk.8 Patients requiring dialysis and those with eGFR ≤20 mL/min/1.73 m2 were excluded from the study. The Revised Cardiac Risk Index was used to assess cardiac risk.23 Evidence-based, perioperative factors associated with perioperative morbidity operative time,24 blood loss/postoperative hematocrit,25 and immediate postoperative temperature,26,27 were collected prospectively.
Postoperative morbidity survey (POMS) was recorded (Table 1) using a validated system.16,18 The POMS was administered by 1 of 2 study nurses to consenting patients on postoperative days 3, 5, 8, and 15. The study nurses were blinded to eGFR data. All data and Revised Cardiac Risk Index analyses were conducted independently of data collectors. POMS criteria were evaluated through direct patient questioning and examination, review of clinical notes and charts, retrieval of data from the hospital clinical information system, and/or consulting with the patient's caregivers. Patients were cared for by the normal attending clinicians who were blinded to the survey results. We also recorded patient's age, gender, measures of preoperative risk, ASA physical status score,28 Physiologic and Operative Severity Score for the enUmeration of Mortality and Morbidity,29 length of hospital stay, and mortality.
Categorical data are summarized as absolute values (percentage). Continuous data are presented as mean ± SD. Characteristics of patients in different eGFR groups were compared using the χ2 test for trend, analysis of variance (depending on the distribution and nature of the data), odds ratio, and hazard ratio (HR). Statistical analyses (NCSS 2004; NCSS, Kaysville, UT) of time to discharge from hospital and time to presence of no morbidity were performed using Kaplan-Meier survival plots (log-rank test). A hierarchical, forward 1-way switching multiple regression analysis model (NCSS 2004) of numeric and categorical variables was performed to assess the association between preoperative (age, gender, ASA grade, Physiologic and Operative Severity Score for the enUmeration of Mortality and Morbidity score, diabetes, hypertension, previous myocardial infarction, angina, heart failure, preoperative white cell count and hematocrit, absolute eGFR) and other evidence-based perioperative factors (length of operation, postoperative hematocrit and temperature) with length of hospital stay. P < 0.05 was considered significant.
The study was powered on the basis of previous studies demonstrating ≥1.5 times increase in risk of all-cause morbidity in CKD patients.1–10 Previous noncardiac surgery studies of morbidity demonstrate approximately 50% of patients to have ≥1 postoperative morbidities after similar orthopedic procedures on postoperative day 5.20,21 From these previous studies, our primary hypothesis was that the absolute incidence of morbidity in those patients with CKD would be >15% higher. Because population data suggested that patients with CKD should constitute approximately 20% of our surgical population,1–10 group sample sizes of at least 389 (eGFR ≥60 mL/min/1.73 m2) and 98 (eGFR ≤60 mL/min/1.73 m2) were required to achieve 80% power to detect a difference between the group proportions of 0.15.30 The test statistic used was the 2-sided Fisher exact test, with the significance level of the test targeted at <0.05.
Patients with normal renal function were more prevalent, younger (P < 0.0001), more likely to be male (P = 0.001), and at less perioperative cardiac risk (P < 0.0001) as defined by the Revised Cardiac Risk Index (Table 2). Perioperative management, including the use of regional analgesia, postoperative core temperature, and intraoperative blood loss (postoperative change in hematocrit) were also similar between groups.
eGFR and Postoperative Morbidity
Postoperative morbidity occurred more frequently in CKD patients (odds ratio 2.1; 95% confidence intervals [CIs]: 1.2–3.7); P < 0.0001; Table 3) throughout the postoperative period (Fig. 1A). CKD was associated with prolonged morbidity (Fig. 1B), as defined by the time required to become morbidity free postoperatively (HR 1.6 [95% CI: 1.2–1.9]; P < 0.0001; log-rank test). Specific patterns of morbidity (Fig. 2A) persisted throughout the postoperative period (Fig. 2B). Within the CKD patient cohort, preoperative eGFR ≤50 mL/min/1.73 m2 was associated with more morbidity (Supplementary Figs. 1 [see Supplementary Digital Content 2, http://links.lww.com/AA/A165] and 2 [see Supplementary Digital Content 3, http://links.lww.com/AA/A166], see Appendix for supplementary figure legends; Supplementary Table 2 [see Supplementary Digital Content 4, http://links.lww.com/AA/A167]).
CKD and the Impact of Morbidity on Length of Hospital Stay
CKD was associated with delayed discharge from hospital (HR 1.4 [95% CI: 1.2–1.7]; P = 0.0001; log rank test; Fig. 3A). Time to discharge was prolonged in all patients who sustained early morbidity on postoperative day 3 (HR 2.4 [95% CI: 1.5–3.6]; P < 0.0001). However, early morbidity in CKD conferred increased hospital stay (Fig. 3B) even compared with those patients with normal renal function who also sustained similar early postoperative complications (HR 1.3 [95% CI: 1.1–1.7]; P = 0.02). Within the CKD patient cohort, preoperative eGFR <50 mL/min/1.73 m2 was associated with prolonged hospital stay (Supplementary Fig. 3, see Supplementary Digital Content 5, http://links.lww.com/AA/A168; see Appendix for supplementary figure legend). Multiple regression analysis identified CKD (P = 0.006) and congestive cardiac failure (P = 0.002) as independent preoperative factors associated with prolonged hospital stay. These 2 factors were not associated with each another (P = 0.47).
A recent meta-analysis of 31 (mostly retrospective) cohort studies demonstrated that CKD conferred increased risk of postoperative death and cardiovascular events in noncardiac surgical patients compared with those with preserved renal function.12 Our data add significant additional information to previous studies, and underscore the importance of CKD in determining postoperative outcomes31 for 4 principal reasons.
First, we have demonstrated that even in moderate-risk surgery, there is a clear relationship between CKD and postoperative morbidity. Large epidemiological studies have shown that perioperative morbidity is associated with dramatic differences in postdischarge life expectancy.13 Second, we provide data that demonstrate the graded relationship between the stage of CKD and prospectively defined postoperative outcomes. Although meta-regression analysis of 5 retrospective studies exclusively conducted in major vascular surgery revealed a graded relationship between severity of CKD and postoperative death,12 the need for further prospective studies in other noncardiac, particularly nonvascular surgical patients, was highlighted. Importantly, unlike our study, previous reports have not used either standardized or population-based definitions to define CKD.22 Third, our data are highly consistent with large-scale, epidemiological studies demonstrating that reductions in GFR, particularly <50 mL/min/1.73 m2, are strongly associated with increased cardiovascular risk and all-cause mortality.1–10 Translating these data into the perioperative environment using prospectively gathered, validated outcome measures is an important contribution because it provides proof-of-principle for the use of preoperative eGFR as a tool to explore surgical morbidity in wider populations. Fourth, physicians can readily assess the severity of CKD using eGFR, which contributes a numerical, universal scale that permits a more refined assessment of postoperative risk, rather than dichotomous information provided by the presence or absence of comorbid conditions. Importantly, traditional risk factors (cardiovascular/cardiac disease) were present in a minority of patients even in eGFR categories at highest risk of postoperative morbidity. Refinement of current definitions of CKD may enhance risk-prediction tools for noncardiac surgery. Current guidelines32 classify a serum creatinine of ≥177 μmol l−1 as an intermediate, but not major, risk factor for postoperative death or cardiovascular complications. The development of novel models should integrate grades of CKD given the robust strength of these data from both population- and surgical-based cohort studies.
Strengths and Limitations of the Study
The chief strength of this study is the prospective collection of postoperative morbidity data using a validated tool18 in a noncardiac surgical population undergoing homogeneous, moderate-risk surgery in whom the impact of chronic renal dysfunction on perioperative outcomes has not undergone prospective, systematic evaluation. We used an internationally adopted classification of CKD22 to explore the relationship between eGFR and postoperative morbidity, because serum creatinine is regarded as an insensitive indicator of renal function.33 The limitations of the factors included in the Modification of Diet in Renal Disease equation, including age, have been debated extensively, although importantly the equation remains accurate in the borderline kidney disease range. The chief limitation in the Modification of Diet in Renal Disease study equation is the underestimation at higher levels of GFR, which does not apply to our study because we have focused our analyses on eGFR <60 mL/min/1.73 m2. Previous studies have used several definitions for CKD, complicating the interpretation between studies and the clinical application of these results. Our results extend previous work substantially, because we have highlighted the importance of renal dysfunction at levels not deemed to be of clinical significance. Our study was an observational cohort study with the inherent limitations of any observational design. Nevertheless, we have conducted the largest prospectively defined and adequately powered study in a very specific, tightly defined surgical subpopulation to explore the relationship between mild to moderate preoperative renal dysfunction and postoperative outcomes. Our assiduous follow-up and rigorous, prospective definition of postoperative morbidity is an important strength. Furthermore, the previously defined prevalence of postoperative morbidity in this particular surgical population17,18 guaranteed adequate statistical power to detect associations for a wide range of levels of eGFR. Although hospitalized patients with impaired renal function sustain more adverse safety events,34 the patterns of morbidity observed in our study were not consistent with such iatrogenic etiologies. For example, we did not observe any abnormal patterns in perioperative factors between eGFR groups that could contribute to adverse safety, such as postoperative hypothermia28,35 or differences in pain therapy. Because no intervention was assessed, these data can only provide associative conclusions. We were restricted in our measures of renal function to eGFR. Cystatin C may be a superior predictor of outcomes, particularly in elderly patients.31 We also have no measure of albuminuria, which may further enhance the predictive value of renal-related pathology.36
Chronic renal disease confers substantially increased risk of postoperative morbidity in homogeneous, elective, moderate-risk orthopedic surgery. Larger studies are required to define the precise contribution that standardized measures of renal function can provide to help refine the stratification of perioperative risk in noncardiac surgery. Preventive perioperative strategies in targeted subpopulations of patients at particularly high risk may be beneficial.
Supported by Academy Medical Sciences/Health Foundation Clinician Scientist Award to GLA, Surgical Outcomes Research Centre, Comprehensive Biomedical Research Centre, University College London Hospitals NHS Trust/University College London and The Centre for Anaesthesia, Pain Management and Critical Care, University College London. This work was undertaken at University College London Hospitals NHS Trust/University College London, which received a proportion of funding from the Department of Health UK NIHR Biomedical Research Centre funding scheme.
GLA helped with study design, data analysis, conduct of study, and manuscript preparation; NM and SC helped with data collection; and MPWG and MGM helped with data collection and manuscript preparation.
The authors acknowledge R. Rivera (intraoperative data collation) and the SOuRCe Surgical Outcomes Resource Centre Investigators (C. Majetowsky, M. Mutch, Y. Zibari, D. Levett, M. Hamilton, M. Emberton, J. Browne, J. Van Der Muellen, F. Haddad, and N. Lees).
1. Meyer KB, Levey AS. Controlling the epidemic of cardiovascular disease in chronic renal disease: report from the National Kidney Foundation Task Force on cardiovascular disease. J Am Soc Nephrol 1998;9:S31–S42
2. Shadman R, Allison MA, Criqui MH. Glomerular filtration rate and N-terminal pro-brain natriuretic peptide as predictors of cardiovascular mortality in vascular patients. J Am Coll Cardiol 2007;49:2172–81
3. Damman K, Navis G, Voors AA, Asselbergs FW, Smilde TD, Cleland JG, van Veldhuisen DJ, Hillege HL. Worsening renal function and prognosis in heart failure: systematic review and meta-analysis. J Card Fail 2007;13:599–608
4. Di AE, Danesh J, Eiriksdottir G, Gudnason V. Renal function and risk of coronary heart disease in general populations: new prospective study and systematic review. PLoS Med 2007; 4:e270
5. Ryan TP, Fisher SG, Elder JL, Winters PC, Beckett W, Tacci J, Sloand JA. Increased cardiovascular risk associated with reduced kidney function. Am J Nephrol 2009;29:620–5
6. Stafford-Smith M. Heart and kidneys: sharing more than just blood. Curr Opin Anaesthesiol 2007;20:65–9
7. Wannamethee SG, Shaper AG, Lowe GD, Lennon L, Rumley A, Whincup PH. Renal function and cardiovascular mortality in elderly men: the role of inflammatory, procoagulant, and endothelial biomarkers. Eur Heart J 2006;27:2975–81
8. Jager A, Kostense PJ, Nijpels G, Dekker JM, Heine RJ, Bouter LM, Donker AJ, Stehouwer CD. Serum homocysteine levels are associated with the development of (micro)albuminuria: the Hoorn study. Arterioscler Thromb Vasc Biol 2001;21:74–81
9. Wannamethee SG, Shaper AG, Lowe GD, Lennon L, Rumley A, Whincup PH. Renal function and cardiovascular mortality in elderly men: the role of inflammatory, procoagulant, and endothelial biomarkers. Eur Heart J 2006;27:2975–81
10. Adams MJ, Irish AB, Watts GF, Oostryck R, Dogra GK. Hypercoagulability in chronic kidney disease is associated with coagulation activation but not endothelial function. Thromb Res 2008;123:374–80
11. Raggi P, Boulay A, Chasan-Taber S, Amin N, Dillon M, Burke SK, Chertow GM. Cardiac calcification in adult hemodialysis patients: a link between end-stage renal disease and cardiovascular disease? J Am Coll Cardiol 2002;39:695–701
12. Mathew A, Devereaux PJ, O'Hare A, Tonelli M, Thiessen-Philbrook H, Nevis IF, Iansavichus AV, Garg AX. Chronic kidney disease and postoperative mortality: a systematic review and meta-analysis. Kidney Int 2008;73:1069–81
13. Khuri SF, Henderson WG, DePalma RG, Mosca C, Healey NA, Kumbhani DJ. Determinants of long-term survival after major surgery and the adverse effect of postoperative complications. Ann Surg 2005;242:326–41
14. Hillis GS, Croal BL, Buchan KG, El-Shafei H, Gibson G, Jeffrey RR, Millar CG, Prescott GJ, Cuthbertson BH. Renal function and outcome from coronary artery bypass grafting: impact on mortality after a 2.3-year follow-up. Circulation 2006;113:1056–62
15. Gibson PH, Croal BL, Cuthbertson BH, Chiwara M, Scott AE, Buchan KG, El-Shafei H, Gibson G, Jeffrey RR, Hillis GS. The relationship between renal function and outcome from heart valve surgery. Am Heart J 2008;156:893–9
16. Bennett-Guerrero E, Welsby I, Dunn TJ, Young LR, Wahl TA, Diers TL, Phillips-Bute BG, Newman MF, Mythen MG. The use of a postoperative morbidity survey to evaluate patients with prolonged hospitalization after routine, moderate-risk, elective surgery. Anesth Analg 1999;89:514–9
17. Ackland GL, Scollay JM, Parks RW, de Beaux I, Mythen MG. Pre-operative high sensitivity C-reactive protein and postoperative outcome in patients undergoing elective orthopaedic surgery. Anaesthesia 2007;62:888–94
18. Grocott MP, Browne JP, Van der Meulen J, Matejowsky C, Mutch M, Hamilton MA, Levett DZ, Emberton M, Haddad FS, Mythen MG. The Postoperative Morbidity Survey was validated and used to describe morbidity after major surgery. J Clin Epidemiol 2007;60:919–28
19. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007;147:573–7
20. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461–70
21. Stevens LA, Coresh J, Feldman HI, Greene T, Lash JP, Nelson RG, Rahman M, Deysher AE, Zhang YL, Schmid CH, Levey AS. Evaluation of the modification of diet in renal disease study equation in a large diverse population. J Am Soc Nephrol 2007;18:2749–57
22. National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002;39:S1–266
23. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, Sugarbaker DJ, Donaldson MC, Poss R, Ho KK, Ludwig LE, Pedan A, Goldman L. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation 1999;100:1043–9
24. Bennett-Guerrero E, Panah MH, Barclay GR, Bodian CA, Winfree WJ, Andres LA, Reich DL, Mythen MG. Decreased endotoxin immunity is associated with greater mortality and/or prolonged hospitalization after surgery. Anesthesiology 2001;94:992–8
25. Hogue CW Jr, Goodnough LT, Monk TG. Perioperative myocardial ischemic episodes are related to hematocrit level in patients undergoing radical prostatectomy. Transfusion 1998;38:924–31
26. Frank SM, Fleisher LA, Breslow MJ, Higgins MS, Olson KF, Kelly S, Beattie C. Perioperative maintenance of normothermia reduces the incidence of morbid cardiac events: a randomized clinical trial. JAMA 1997;277:1127–34
27. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. Study of Wound Infection and Temperature Group. N Engl J Med 1996;334:1209–15
28. American Society of Anesthesiologists. New classification of physical status. Anesthesiology 1963;24:111
29. Copeland GP, Jones D, Walters M. POSSUM: a scoring system for surgical audit. Br J Surg 1991;78:355–60
30. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988
31. Byers J, Sladen RN. Renal function and dysfunction. Curr Opin Anaesthesiol 2001;14:699–706
32. Fleisher LA, Beckman JA, Brown KA, Calkins H, Chaikof EL, Fleischmann KE, Freeman WK, Froehlich JB, Kasper EK, Kersten JR, Riegel B, Robb JF, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Antman EM, Buller CE, Creager MA, Ettinger SM, Faxon DP, Fuster V, Halperin JL, Hiratzka LF, Hunt SA, Lytle BW, Nishimura R, Ornato JP, Page RL, Riegel B, Tarkington LG, Yancy CW. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: executive summary—a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to revise the 2002 guidelines on perioperative cardiovascular evaluation for noncardiac surgery) developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. J Am Coll Cardiol 2007; 50:1707–32
33. Astor BC, Levey AS, Stevens LA, Van Lente F, Selvin E, Coresh J. Method of glomerular filtration rate estimation affects prediction of mortality risk. J Am Soc Nephrol 2009;20:2214–22
34. Seliger SL, Zhan M, Hsu VD, Walker LD, Fink JC. Chronic kidney disease adversely influences patient safety. J Am Soc Nephrol 2008;19:2414–9
35. Slotman GJ, Jed EH, Burchard KW. Adverse effects of hypothermia in postoperative patients. Am J Surg 1985;149:495–501
36. Astor BC, Hallan SI, Miller ER III, Yeung E, Coresh J. Glomerular filtration rate, albuminuria, and risk of cardiovascular and all-cause mortality in the US population. Am J Epidemiol 2008;167:1226–34
APPENDIX: SUPPLEMENTARY FIGURE LEGENDS
Supplementary Figure 1. Time to become free of postoperative morbidity according to severity of chronic kidney disease. Patients with preoperative estimated glomerular filtration rate (eGFR) <50 mL/min/1.73 m2 experienced slower resolution of morbidity.
Supplementary Figure 2. Severity of chronic kidney disease and specific postoperative morbidities. Number of specific postoperative complications at any point postoperatively according to stratification by preoperative estimated glomerular filtration rate (GFR). Asterisk denotes increased risk (determined by significant odds ratio [95% confidence intervals]) for developing morbidity (P < 0.05) compared with preoperative normal renal function.
Supplementary Figure 3. Preoperative estimated glomerular filtration rate (eGFR) and time to hospital discharge. A, Kaplan-Meier plot depicting prolonged time to discharge in patients with preoperative eGFR <50 mL/min/1.73 m2 (P < 0.001). B, Kaplan-Meier plot demonstrating that prolonged time to discharge occurs in chronic kidney disease patients with preoperative eGFR <50 mL/min/1.73 m2 (P = 0.01; log-rank test).