Whether prognosis differs in acute renal failure (ARF) after coronary artery bypass grafting (CABG) in patients with and without recovery of renal function is not known. We studied patients who had CABG at Duke University Medical Center (1995 to 2008). ARF was defined as an increase in peak creatinine ≥50% after CABG or ≥0.7 mg/dl above baseline or need for new dialysis. Patients were categorized into 3 groups: (1) no ARF after CABG, (2) ARF after CABG and completely recovered renal function at day 7 (return of creatinine to no higher than baseline and no dialysis), or (3) ARF after CABG with no recovery of renal function at day 7 (creatinine no higher than baseline or new dialysis). Main outcome measurement was risk-adjusted long-term mortality (excluding death ≤7 days). ARF after CABG occurred in 2,083 of 10,415 patients (20%) and completely recovered in 703 (33.7%). Risk-adjusted mortality was highest in patients with ARF without recovery of renal function (hazard ratios 1.47, 95% confidence interval 1.34 to 1.62) and intermediate in those with ARF but completely recovered renal function (hazard ratios 1.21, 95% confidence interval 1.07 to 1.37, referent no-ARF group). Mortality was lower in patients with ARF compared to those without complete recovery of renal function (p = 0.0083). In conclusion, in patients with ARF after CABG, complete recovery of renal function was associated with significantly lower long-term mortality compared to those without such recovery, although this was significantly higher than in those without ARF. Thus, major emphasis should be on prevention of ARF in patients undergoing CABG.
Acute renal failure (ARF) after coronary artery bypass graft surgery (CABG) is common, with an incidence of 3% to 30%, depending on the definition used. Compared to patients without ARF, those who develop this complication after CABG have been shown to have higher short-term morbidity and mortality. However, it is not known if the prognosis is different in patients with ARF in whom renal function recovers compared to those without recovery of renal function before discharge. The goal of the present investigation was to evaluate the outcomes of patients after CABG with and without recovery in postoperative ARF.
Methods
Using the Duke Cardiovascular Disease Databank, we identified patients who underwent CABG from January 1, 1995 to June 30, 2008 (n = 12,952). We excluded patients with previous dialysis or preoperative creatinine level >2 mg/dl (n = 590), those undergoing salvage CABG (n = 14), those without pre- and ≥2 postoperative creatinine values (n = 1,554), and those who died ≤7 days after CABG (n = 379).
Acute kidney injury was defined as an increase in peak creatinine ≥50% after CABG or ≥0.7 mg/dl above baseline values or need for new dialysis. Complete recovery of renal function was defined as a decrease in peak creatinine no higher than baseline values after CABG and no new dialysis by postoperative day 7 (or at discharge if this occurred before postoperative day 7). No recovery was defined as a creatinine level no higher than baseline values or new dialysis by postoperative day 7 (or at discharge if this occurred before postoperative day 7). Hypertension, diabetes mellitus, dyslipidemia, peripheral vascular disease, and presenting heart failure were ascertained as reported on the data collection form as present or absent. After their index CABG, patients were followed at 6 months, 1 year, and annually thereafter for mortality status and other adverse events by telephone contact, mailed questionnaire, and National Death Index search. Methods used by the Duke Cardiovascular Disease Databank have been described previously.
Summary statistics are presented as frequencies and percentages for categorical variables and as medians with interquartile ranges (25th to 75th percentiles) for continuous variables. Comparisons among the 3 groups (i.e., those without ARF and those with ARF with and without recovery of renal function at discharge) were made using chi-square test for categorical variables and Kruskal-Wallis test for continuous variables. Previously developed Cox proportional hazards model for long-term mortality was used to derive estimate of adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) of long-term mortality.
We performed several sensitivity analyses. Thus, we evaluated the impact of recovery of renal function in patients undergoing elective CABG and in those undergoing isolated CABG (without concomitant valve or other cardiac surgeries). We also considered discharge creatinine to reflect complete or no recovery in those discharged alive to account for patients who may have had recovery of renal function >7 days. We then defined ARF using the risk of renal dysfunction, injury to the kidney, failure of kidney function, loss of kidney function and end-stage kidney disease (RIFLE) classification (≥50% increase in creatinine above baseline value) and compared outcomes of patients without ARF and those with ARF with and without complete recovery of renal function. We divided the group with no recovery of renal function group into 2 subgroups: (1) those with partial recovery of renal function (creatinine decrease >50% of peak postoperative creatinine or decrease >0.7 mg/dl in peak postoperative creatinine, but no higher than baseline values, and no new dialysis at day 7 or at discharge if sooner than day 7), and (2) those without any recovery (predischarge creatinine decrease ≤50% of peak postoperative creatinine or decrease ≤0.7 mg/dl in peak postoperative creatinine, but no higher than baseline values, or need for new dialysis at day 7 or at discharge if sooner than day 7) and compared risk-adjusted long-term mortality in these groups. All analyses were performed using SAS statistical software (SAS Institute, Cary, North Carolina).
Results
Of the 10,415 patients undergoing CABG who were eligible for this study, ARF occurred in 2,083 subjects (20%). Of patients with ARF, renal function recovered to baseline levels in 703 patients (33.7%). Median follow-up for the entire group was 7.0 years (25th to 75th percentiles 3.8 to 10.0). Table 1 presents distributions of baseline clinical and angiographic features of the 3 groups of patients. Compared to patients without ARF, those with this event were significantly older, were more likely to be nonwhite, and had significantly higher co-morbid conditions. Hemodynamic instability, emergency or urgent CABG, and longer pump and cross-clamp times were more common in patients with ARF compared to those without this complication. Left ventricular ejection fraction was lower in patients with ARF.
| Characteristic | Overall | No ARF | ARF With Recovery | ARF With No Recovery | p Value ⁎ | p Value † |
|---|---|---|---|---|---|---|
| Number of patients | 10,415 (100%) | 8,332 (80.0%) | 703 (6.7%) | 1,380 (13.3%) | ||
| Age (years) ‡ | 65 (57–73) | 65 (56–72) | 66 (58–73) | 68 (59–75) | <0.0001 | 0.0033 |
| Female gender | 3,274 (31%) | 2,617 (31%) | 240 (34%) | 417 (30%) | 0.1885 | 0.0685 |
| White race | 7,840 (79%) | 6,367 (80%) | 504 (76%) | 969 (74%) | <0.0001 | 0.3794 |
| Body mass index (kg/m 2 ) | 27.8 (25–31) | 27.7 (25–31) | 28.7 (25–33) | 28.7 (26–33) | <0.0001 | 0.2664 |
| Hypertension | 7,711 (74%) | 6,028 (72%) | 565 (80%) | 1,118 (81%) | <0.0001 | 0.7239 |
| Diabetes mellitus | 3,584 (34%) | 2,703 (32%) | 297 (42%) | 584 (42%) | <0.0001 | 0.9751 |
| Current/former smoker | 5,866 (56%) | 4,693 (56%) | 392 (56%) | 781 (57%) | 0.9364 | 0.7170 |
| Dyslipidemia § | 6,381 (61%) | 5,086 (61%) | 420 (60%) | 875 (63%) | 0.1716 | 0.1032 |
| Previous myocardial infarction | 4,202 (40%) | 3,310 (40%) | 294 (42%) | 598 (43%) | 0.0290 | 0.5095 |
| Chronic lung disease | 1,330 (13%) | 1,008 (12%) | 106 (15%) | 216 (16%) | 0.0002 | 0.7319 |
| Previous percutaneous coronary intervention | 1,565 (15%) | 1,254 (15%) | 113 (16%) | 198 (14%) | 0.5753 | 0.2959 |
| Previous coronary bypass | 458 (4.4%) | 336 (4.0%) | 25 (3.6%) | 97 (7.0%) | <0.0001 | 0.0014 |
| Previous congestive heart failure | 2,667 (26%) | 1,946 (23%) | 252 (36%) | 469 (34%) | <0.0001 | 0.3986 |
| Cerebrovascular disease | 1,362 (13%) | 1,023 (12%) | 98 (14%) | 241 (18%) | <0.0001 | 0.0394 |
| Peripheral arterial disease | 1,701 (16%) | 1,288 (16%) | 135 (19%) | 278 (20%) | <0.0001 | 0.6103 |
| Presenting characteristics | ||||||
| Heart rate (beats/min) ‡ | 70 (62–81) | 70 (61–80) | 75 (64–85) | 71 (62–83) | <0.0001 | 0.0223 |
| Systolic blood pressure (mm Hg) ‡ | 136 (118–152) | 135 (118–151) | 136 (119–157) | 138 (120–155) | 0.0028 | 0.9751 |
| New York Heart Association class ≥III | 1,680 (16%) | 1,224 (15%) | 173 (25%) | 283 (21%) | <0.0001 | 0.9323 |
| Cardiogenic shock | 242 (2.3%) | 166 (2.0%) | 18 (2.6%) | 58 (4.2%) | <0.0001 | 0.0587 |
| Status of surgery | <0.0001 | <0.0001 | ||||
| Emergency | 291 (2.8%) | 196 (2.4%) | 17 (2.4%) | 78 (5.7%) | ||
| Urgent | 4,376 (42%) | 3,441 (41%) | 236 (34%) | 699 (51%) | ||
| Elective | 5,730 (55%) | 4,679 (56%) | 450 (64%) | 601 (44%) | ||
| Perfusion time (minutes) ‡ | 112 (88–142) | 110 (86–139) | 117 (91–149) | 123 (95–167) | <0.0001 | 0.0003 |
| Cross-clamp time (minutes) ‡ | 60 (42–82) | 59 (41–80) | 62 (44–85) | 68 (46–96) | <0.0001 | 0.0071 |
| Internal mammary artery bypass grafting | 9,318 (90%) | 7,487 (90%) | 632 (90%) | 1,199 (90%) | 0.0036 | 0.0459 |
| Coronary artery bypass grafting + valve surgery | 1,175 (11%) | 815 (10%) | 116 (17%) | 244 (18%) | <0.0001 | 0.5005 |
| Angiographic features | ||||||
| Left ventricular ejection fraction (%) | 51 (41–64) | 52 (43–65) | 50 (39–61) | 48 (37–62) | <0.0001 | 0.1758 |
| Diseased coronary arteries ≥2 | 9,485 (91%) | 7,567 (91%) | 644 (92%) | 1,274 (92%) | 0.1700 | 0.5697 |
| Left main coronary artery disease (>50% narrowing) | 2,809 (27%) | 2,235 (27%) | 173 (25%) | 401 (29%) | 0.0768 | 0.0316 |
| Laboratory data | ||||||
| Hemoglobin (gm/dl) ‡ | 13.2 (12–14) | 13.3 (12–14) | 12.7 (11–14) | 12.6 (11–14) | <0.0001 | 0.4424 |
| Preoperative serum creatinine (mg/dl) ‡ | 1.0 (0.9–1.2) | 1.0 (0.9–1.2) | 1.1 (1.0–1.5) | 1.1 (0.9–1.3) | <0.0001 | <0.0001 |
| Postoperative peak serum creatinine (mg/dl) ‡ | 1.2 (1.0–1.6) | 1.2 (1.0–1.4) | 2.0 (1.6–2.3) | 2.0 (1.6–2.6) | <0.0001 | 0.0029 |
| Discharge serum creatinine (mg/dl) ‡ | 1.0 (0.9–1.2) | 1.0 (0.9–1.2) | 1.1 (0.9–1.3) | 1.4 (1.1–1.9) | <0.0001 | <0.0001 |
† For difference in acute renal failure with and without complete recovery.
‡ The p values for these continuous variables determined using Kruskal-Wallis test.
§ History of increased low-density lipoprotein, triglycerides, and/or low high-density lipoprotein or use of lipid-lowering agents.
In contrast, baseline variables were similar in patients with ARF who recovered completely compared to those in whom renal function did not recover completely, with a few exceptions. Patients with complete recovery of renal function were younger and less likely to have cerebrovascular disease, previous CABG, or higher New York Heart Association functional class. They had higher serum creatinine values before CABG but lower values after CABG. They were also less likely to have left main coronary artery disease or to have undergone emergency or urgent surgery. Cross-clamp and perfusion times were shorter in this cohort.
Table 2 lists in-hospital events and 30-day mortality in the 3 groups. Most in-hospital clinical events and median length of stay were increased in patients with ARF.
| Event | Overall | No ARF | ARF With Recovery | ARF With No Recovery | p Value ⁎ | p Value † |
|---|---|---|---|---|---|---|
| Number | 10,415 | 8,332 (80.0%) | 703 (6.7%) | 1,380 (13.3%) | ||
| Death 8–30 days | 68 (0.7%) | 49 (0.6%) | 8 (1.1%) | 11 (0.8%) | 0.1710 | 0.4687 |
| Permanent stroke | 118 (1.1%) | 86 (1.0%) | 6 (0.9%) | 26 (1.9%) | 0.0166 | 0.0889 |
| Perioperative myocardial infarction | 11 (0.1%) | 5 (0.1%) | 1 (0.1%) | 5 (0.4%) | 0.0057 | 0.6704 |
| Prolonged ventilation | 458 (4.4%) | 289 (3.5%) | 33 (4.7%) | 136 (9.9%) | <0.0001 | <0.0001 |
| Pneumonia | 116 (1.1%) | 60 (0.7%) | 14 (2.0%) | 42 (3.0%) | <0.0001 | 0.1604 |
| Cardiac tamponade | 216 (2.1%) | 142 (1.7%) | 23 (3.3%) | 51 (3.7%) | <0.0001 | 0.6211 |
| Any blood product use | 4,794 (46%) | 3,644 (44%) | 370 (53%) | 780 (57%) | <0.0001 | 0.0914 |
| Postoperative intra-aortic balloon pump use | 86 (0.8%) | 43 (0.5%) | 4 (0.6%) | 39 (2.8%) | <0.0001 | 0.0003 |
| Deep sternal wound infection | 77 (0.7%) | 45 (0.5%) | 10 (1.4%) | 22 (1.6%) | <0.0001 | 0.8521 |
| Septicemia | 16 (0.2%) | 3 (0.1%) | 1 (0.1%) | 12 (0.9%) | <0.0001 | 0.0725 |
| Postoperative length of stay (days) ‡ | 5 (4–7) | 5 (4–7) | 7 (5–9) | 8 (6–13) | <0.0001 | <0.0001 |
† For difference in acute renal failure with and without recovery.
‡ Values for these continuous variables determined using Kruskal-Wallis test.
In-hospital outcomes and 30-day death were similar in the ARF groups with and without complete recovery of renal function. However, prolonged ventilation and postoperative intra-aortic balloon pump were used more frequently in patients with ARF who did not have complete recovery of renal function. Long-term outcomes were significantly worse in patients with ARF without complete recovery of renal function ( Table 3 ). Compared to patients without ARF, long-term mortality was significantly higher in patients with ARF without complete recovery of renal function (HR 1.47, 95% CI 1.34 to 1.62) and intermediate for those with ARF but completely recovered renal function (HR 1.21, 95% CI 1.07 to 1.37, referent no-ARF group; Figure 1 ).
| Characteristic | HR | 95% CI | p Value |
|---|---|---|---|
| Acute renal failure with complete recovery ⁎ | |||
| Death or rehospitalization | 1.22 | 1.12–1.33 | <0.0001 |
| Death or repeat revascularization | 1.30 | 1.16–1.45 | <0.0001 |
| Death or myocardial infarction | 1.37 | 1.23–1.54 | <0.0001 |
| Death | 1.48 | 1.31–1.67 | <0.0001 |
| Acute renal failure with no recovery ⁎ | |||
| Death or rehospitalization | 1.41 | 1.33–1.50 | <0.0001 |
| Death or repeat revascularization | 1.52 | 1.40–1.66 | <0.0001 |
| Death or myocardial infarction | 1.73 | 1.59–1.89 | <0.0001 |
| Death | 1.78 | 1.62–1.95 | <0.0001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
