Intraoperative Hypotension Is Not Correlated With Acute Kidney Injury During Spinal Fusion Surgery

Discussion

This study sought to determine whether IOH during spinal fusion surgery is independently associated with adverse outcomes such as AKI, myocardial injury (as measured by troponin leak), and postoperative ileus. IOH was characterized as 10 or more cumulative minutes with an MAP of less than 65 mm Hg. AKI was characterized by the ratio of preoperative to postoperative serum creatinine and was staged based on the KDIGO criteria. Using these definitions, we failed to find any independent association between IOH during spinal fusions and AKI, troponin leak, or postoperative ileus. To our knowledge, this is the first study investigating the relationship between IOH and organ injury in spine surgery.

Current literature strongly indicates that IOH is correlated with AKI, though these observational studies have not been able to establish a causative relationship. The hypothesized etiology for AKI after occurrence of IOH stems from previous human13 and animal14,15 studies, which suggest that the lower limit of renal blood flow autoregulation is at an MAP of 50–75 mm Hg, after which blood flow becomes pressure-dependent. In the event of IOH, where MAP reaches the lower threshold of the autoregulation curve, ischemic injury can damage the kidney resulting in AKI. While these thresholds are individual rather than empiric, preventing the definitive conclusion that IOH leads to AKI,4 numerous previous studies have suggested this relationship. For example, Salmasi et al7 found that hypotensive events less than 65 mm Hg were associated with higher odds of both AKI and myocardial injury. This association appears to be time sensitive as well; a systematic review of 42 studies regarding IOH and organ injury found that the risk of organ injury seemed to start at blood pressures less than 80 mm Hg if durations were longer than 10 minutes.16

In contrast to this body of literature, our data do not reveal an independent association between IOH and AKI. Furthermore, the lack of association between IOH and AKI was independent on the duration of IOH that the patient experienced. Several factors could contribute to our findings being inconsistent with previous studies, including the nature of spine procedures and characteristics inherent to the neurosurgical patient population.

One possible reason for the lack of association between IOH and organ injury is the general health of this patient cohort. Mathis et al reviewed more than 130,000 cases of major noncardiac surgeries, using factors such as surgery type, eGFR, and American Society of Anesthesiology (ASA) physical status classification to stratify patients into risk quartiles. Interestingly, they found that low-risk patients could tolerate MAP values as low as 50 mm Hg with no association to AKI.17 While ASA class was not explicitly measured in our patient cohort, studies have shown that the proportion of lumbar fusion patients with ASA class I or II can range as high as 94.8% to 99.6%, leaving very few patients in the higher-risk ASA classes.18,19 It is plausible that our patient cohort comprised a greater proportion of low ASA class patients, and thus, were less sensitive to AKI after experiencing IOH as compared with surgical patient populations with higher average ASA classes.

Different surgical protocols may also affect the relationship between IOH and AKI. It has been previously suggested that the association between IOH and AKI may not be due to a direct effect from IOH itself but rather how IOH is managed in the operating room.4 For instance, administration of high volumes of normal saline (NS; 0.9% sodium chloride) has been shown to cause increased rates of renal dysfunction in patients when compared with the use of balanced crystalloids.20 It is possible that NS administration acts as a confounding variable between IOH and AKI, creating different relationships in surgeries that require varying levels of fluid replacement. However, the effect of NS administration on renal function has been primarily studied in patients who are critically ill or in septic shock.21 Further research is required to understand the effect of unbalanced fluids in the setting of elective surgery and whether such a confounder could have a contributory role in the discrepancy between our findings and current literature.

While we offer possible explanations for our data, it is important to note that more investigation is necessary to fully determine the extent to which IOH and AKI are related. While our analysis indicates that AKI may not be as pertinent as an issue in spinal fusion patients who experienced IOH as previously believed, standard surgical protocols should be followed to avoid and treat IOH; more research must be done to completely understand this phenomenon.

Our study did find correlations between IOH and various secondary outcomes measured. For example, IOH was significantly associated with the use of vasopressors; this finding serves to validate our study, as hypotension is an indication for intraoperative vasopressor administration. Additionally, EBL was associated with total IOH duration but not the binary variable of IOH occurrence. Again, this is not surprising because a longer duration of IOH, rather than a transient IOH period, may indicate a higher quantity of blood lost. Last, preoperative serum creatinine correlated to LOS, possibly indicating that preoperative creatinine is a marker of overall patient health; sicker patients may require a longer recovery after surgery, an interpretation that has been proposed in previous literature.22,23

Our study had several strengths. Our use of an arterial line to measure MAP provided us with a more accurate and physiological evaluation of blood pressure than using an external cuff to measure systolic blood pressure.24 Additionally, our choice to use a partial correlation analysis allowed for the evaluation of the relationship between IOH and our primary outcomes while controlling for other variables, such as age, sex, preoperative use of angiotensin-converting enzyme inhibitors, etc.

This study also had limitations. First, IOH was defined using a threshold of MAP <65 mm Hg; thus quantifying the degree of hypotension (eg, if MAP drops to 55 mm Hg) was not possible. Future studies should attempt to investigate the relationship between AKI and IOH with hypotension as a continuous variable, as previous literature has identified a graded relationship between the extent of IOH in a patient and the risk of organ injury.16 We also viewed the time that a patient was hypotensive as a cumulative measure (ie, a patient hypotensive for 1 minute at 10 separate timepoints is equivalent to a patient who was continuously hypotensive for 10 minutes). This could have diluted the relationship between IOH and organ injury as hypotension for just a short amount of time with recovery periods in between allow for organ reperfusion and potentially minimize acute damage. Last, our findings may also caution against future study designs that group multiple types of major surgeries into the same analysis, as factors unique to different types of surgery and patient populations may impact the measured association between IOH and AKI.