J Neurosurg 120:1340–1348, 2014 ©AANS, 2014
Increased risk of acute kidney injury associated with higher infusion rate of mannitol in patients with intracranial hemorrhage Clinical article Min Young Kim, M.D.,1 Ji Hyeon Park, M.D., 2 Na Ree Kang, M.D.,1 Hye Ryoun Jang, M.D., 2 Jung Eun Lee, M.D., 2 Wooseong Huh, M.D., 2 Yoon-Goo Kim, M.D., 2 Dae Joong Kim, M.D., 2 Seung-Chyul Hong, M.D., 3 Jong-Soo Kim, M.D., 3 and Ha Young Oh, M.D. 2 Department of Internal Medicine, Seoul Medical Center; and 2Division of Nephrology, Department of Medicine, and 3Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 1
Object. Mannitol, an osmotic agent used to decrease intracranial pressure, can cause acute kidney injury (AKI). The objectives of this study were to assess the impact of mannitol on the incidence and severity of AKI and to identify risk factors and outcome for AKI in patients with intracranial hemorrhage (ICH). Methods. The authors retrospectively evaluated 153 adult patients who received mannitol infusion after ICH between January 2005 and December 2009 in the neurosurgical intensive care unit. Multivariate analysis was used to evaluate the risk factors for AKI after ICH. Based on the odds ratio, weighted scores were assigned to predictors of AKI. Results. The overall incidence of AKI among study participants was 10.5% (n = 16). Acute kidney injury occurred more frequently in patients who received mannitol infusion at a rate ≥ 1.34 g/kg/day than it did in patients who received mannitol infusion at a rate < 1.34 g/kg/day. A higher mannitol infusion rate was associated with more severe AKI. Independent risk factors for AKI were mannitol infusion rate ≥ 1.34 g/kg/day, age ≥ 70 years, diastolic blood pressure (DBP) ≥ 110 mm Hg, and glomerular filtration rate < 60 ml/min/1.73 m2. The authors developed a risk model for AKI, wherein patients with a higher risk score showed a graded association with a higher incidence of AKI. Conclusions. The incidence of AKI following mannitol infusion in patients with ICH was 10.5%. A higher mannitol infusion rate was associated with more frequent and more severe AKI. Additionally, age ≥ 70 years, DBP ≥ 110 mm Hg, and established renal dysfunction before starting mannitol therapy were associated with development of AKI. (http://thejns.org/doi/abs/10.3171/2013.12.JNS13888)
Key Words • acute kidney injury • intracranial hemorrhage • mannitol • vascular disorders
I
hemorrhage (ICH) is a severe disease with a mortality rate approaching 50% and nearly 50% of survivors left with cognitive deficits or motor dysfunction.4,10,29,34,38,41 Patients with ICH suffer from additional brain injury caused by increased intracranial pressure (ICP), which requires prompt intervention to prevent devastating neurological outcomes. ntracranial
Abbreviations used in this paper: AKI = acute kidney injury; AKIN = Acute Kidney Injury Network; AUC = area under the curve; DBP = diastolic blood pressure; GFR = glomerular filtration rate; ICH = intracranial hemorrhage; ICP = intracranial pressure; IQR = interquartile range; ROC = receiver operating characteristic; SBP = systolic blood pressure.
1340
Mannitol, a hyperosmolar agent that causes osmotic diuresis, is considered useful in the treatment of increased ICP.33 The effect of mannitol therapy is dose dependent, with higher doses providing a longer-lasting reduction in ICP.37,40 However, marked hyperosmolality may cause acute kidney injury (AKI).3,13 Previous studies have shown that extracerebral organ system dysfunctions are related to unfavorable neurological outcomes and high mortality rates in patients with brain injury and hemorrhage.16,36 In particular, AKI is associated with worse outcome in patients with brain injury.22,42 However, there is little information published regarding the incidence, risk factors, and outcome of mannitol-induced AKI. In previous case reports, AKI occurred at even small J Neurosurg / Volume 120 / June 2014
Mannitol-induced AKI (< 400 g) doses of mannitol.2,13,35,39 Pérez-Pérez et al. concluded that AKI is related to the rate of infusion rather than the total dose of mannitol.32 These results suggest that AKI may be associated with the infusion rate of mannitol rather than the dosage. The Acute Kidney Injury Network (AKIN) proposed a modified diagnostic staging system for AKI that focuses on small changes in serum creatinine.28 The AKIN classification has been validated in intensive care units and has been found to predict in-hospital mortality, the need for renal replacement therapy, and prolonged hospital stay.1 Thus, the present study was undertaken to evaluate the incidence of and associated risk factors for AKI resulting from mannitol therapy in patients treated for ICH and to determine if the infusion rate is associated with AKI, as defined by the AKIN classification.
in a relatively consistent manner. Serum electrolytes and osmolality were measured before each dose of mannitol, at least once to several times a day as needed. If serum osmolality exceeded 285 mOsm/kg, 100 ml of normal saline was infused. If serum osmolality exceeded 315 mOsm/kg, 100 ml of 0.45% normal saline was infused. The infusion rate was calculated as follows: total dose of mannitol (g)/body weight (kg)/day. A receiver operating characteristic (ROC) curve was applied to determine the cutoff mannitol infusion rate required to predict AKI. The ROC curve analysis resulted in a cutoff of 1.34 g/kg/day for AKI probability. Using this cutoff, the prediction rule showed a 75.0% sensitivity, a 65.7% specificity, and an area under the curve (AUC) of 0.71 (p = 0.007, 95% CI 0.58–0.84).
Methods
Continuous variables are expressed as the median with the interquartile range (IQR) or as the mean ± SD. Categorical variables are expressed as counts (percentages). Between-group comparisons of body weight, serum osmolality, and creatinine were made using the Student t-test or the Mann-Whitney U-test, as appropriate. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for the development of AKI in patients with ICH. Variables with a p value < 0.05 by univariate analysis were entered into multivariate analysis. Odds ratios and 95% CIs were calculated, and, based on the odds ratio, weighted scores were assigned to each of the AKI predictors. We then evaluated the incidence of AKI according to the score value. The evaluation of each AKI predictor produced an ROC curve; we used the AUC to assess the power of discrimination between patients with and without AKI. All analyses were performed using SPSS version 12.0 (SPSS, Inc.), and p values < 0.05 were considered significant.
We retrospectively studied 537 consecutive adult patients who received mannitol infusion after ICH between January 2005 and December 2009 in the neurosurgical intensive care unit. We excluded 384 patients with tumors (n = 32); renal failure requiring dialysis (n = 3); no measurement of serum creatinine before, during, and after mannitol therapy (n = 179); an administration interval of mannitol more than 2 days (n = 93); AKI (an increase in serum creatinine 0.3 mg/dl from baseline) on admission (n = 14); and patients transferred from outside hospitals during therapy (n = 63). The remaining 153 patients were included in the study. Patient data were collected from electronic medical records and databases. The demographic data and clinical data collected were age, sex, body weight, systolic blood pressure (SBP), diastolic blood pressure (DBP), and status at time of admission for all of the following conditions: diabetes mellitus, hypertension, liver disease, cerebrovascular accident, ischemic heart disease, Glasgow Coma Scale score, and type of ICH. Administration of mannitol, either alone or concurrently with other drugs (specifically furosemide, nonsteroidal antiinflammatory drugs, osmotic agents, and potentially nephrotoxic antibiotics), radiocontrast agents, inotropic agents, barbiturates, decompressive craniectomy, and adverse events such as myocardial infarction and infection were also investigated. We examined the laboratory results for serum hemoglobin, glucose, urea, creatinine, sodium, osmolality, and glomerular filtration rate (GFR). The GFR was estimated using the 4-variable Modification in Diet and Renal Disease equation.21 The study was approved by our institutional review board. The primary outcome was the occurrence of AKI, defined as a ≥ 0.3-mg/dl increase in serum creatinine level above baseline or a change of ≥ 50%, according to the AKIN classification. The severity of AKI was classified according to the AKIN system as follows: Stage 1 (increase in serum creatinine level by ≥ 50% or ≥ 0.3 mg/dl), Stage 2 (increase in serum creatinine level by > 100%), and Stage 3 (increase in serum creatinine level by > 200% or by ≥ 0.5 mg/dl with serum creatinine level of ≥ 4.0 mg/dl).28 Mannitol was administered intravenously by bolus J Neurosurg / Volume 120 / June 2014
Statistical Analysis
Results
Patient characteristics are summarized in Tables 1 and 2. The mean age of patients was 55 ± 14 years, and 77 patients (50.3%) were male (Table 1). Of the 16 patients (10.5%) in whom AKI developed, 1 (6.3%) received a total of 3 hemodialysis sessions. Complete recovery (return to baseline serum creatinine level) was observed in 11 of 16 patients, including the patient who underwent hemodialysis, a median of 4 days after hospitalization (IQR 2.5–10.5 days). The mortality rate was approximately 3.5 times higher in patients with AKI than in those without (12.5% vs 3.6%, p = 0.133). Table 3 lists the variables significantly associated with the development of AKI in patients who received mannitol in univariate analysis. The significant variables associated with AKI were age ≥ 70 years, DBP ≥ 110 mm Hg, GFR < 60 ml/min/1.73 m2, and a rate of mannitol infusion ≥ 1.34 g/kg/day in multivariate analysis (Table 4). More patients who received a mannitol infusion rate of ≥ 1.34 g/kg/day developed AKI and presented predominantly with severe AKI than those who received a lower rate (trend, p = 0.002, Fig. 1). Figure 2 shows the changes in body weight, serum 1341
M. Y. Kim et al. TABLE 1: Baseline characteristics of patients on admission* Variable
Value
no. of patients mean age in yrs no. of males mean body weight in kg mean SBP in mm Hg mean DBP in mm Hg median Glasgow Coma Scale score medical history hypertension diabetes mellitus ischemic heart disease cerebrovascular accident type of ICH intracerebral hemorrhage subarachnoid hemorrhage intracerebral & subarachnoid hemorrhage subdural or epidural hemorrhage laboratory data mean hemoglobin in g/dl mean GFR in ml/min/1.73 m2 median serum sodium in mmol/L median serum osmolality in mOsm/kg
153 55 ± 14 77 (50.3%) 62.4 ± 10.1 151 ± 33 86 ± 18 14 (9–15) 64 (42.1%) 11 (7.2%) 7 (4.6%) 16 (10.5%) 57 (36.3%) 71 (45.2%) 5 (3.2%) 20 (12.7%) 13.7 ± 1.7 95.0 ± 22.9 139 (138–141) 297 (292–304)
* Continuous variables are expressed as the median (IQR) or as the mean ± SD. Categorical variables are expressed as counts (%).
osmolality, and serum creatinine level during hospitalization in patients with and without AKI. The body weight of patients with AKI tended to increase until 2 days after stopping mannitol and, on discharge, tended to be lower in comparison with that of patients without AKI. However, the difference between groups was not statistically significant. There was a significant difference in serum osmolality and creatinine level between patients with AKI and those without AKI. A risk model was constructed using the predictors of AKI (Table 5). In the present study, risk score levels ranged between 0 and 4 points. The relationship between risk score and AKI incidence is graded: the higher the score, the stronger the association with AKI (Fig. 3). Scores were applied to predict risk for AKI after ICH. The area under the ROC curve was 0.917 (95% CI 0.851– 0.983, Fig. 4). In case of a risk score of 3, positive and negative predictive values were greater than 90% (Table 6).
Discussion
In this study, we evaluated the incidence of and risk factors for AKI in patients treated with mannitol for ICH. Of 153 patients, 16 (10.5%) developed AKI. Risk factors for AKI were identified as age ≥ 70 years, DBP ≥ 110 mm Hg, GFR < 60 ml/min/1.73 m2, and a rate of mannitol infusion ≥ 1.34 g/kg/day. The incidence of AKI increased in patients with a higher risk score. 1342
TABLE 2: Characteristics of 153 patients during hospitalization* Variable treatment mannitol† median duration of Tx in days mean maximum dose in g/kg/day mean total dose in g/kg median infusion rate in g/kg/day glycerol† starch† furosemide† ACEI/ARB† nephrotoxic antibiotics†‡ NSAIDs† radiocontrast agent† inotropics† barbiturate† decompressive craniectomy† mean operative duration in hrs§ median peak serum osmolality in mOsm/kg myocardial infarction†¶ infectious disease†¶
Value 153 (100%) 8 (5–11) 1.46 ± 0.37 9.41 ± 6.02 1.26 (1.07–1.48) 49 (32%) 101 (66%) 45 (29.4%) 25 (16.3%) 39 (25.5%) 111 (72.5%) 118 (77.1%) 40 (26.1%) 12 (7.8%) 17 (11.1%) 3.9 ± 1.1 312 (306–320) 1 (0.6%) 27 (17.6%)
* Mean values are presented ± SD; median values are presented with the IQR. ACEI/ARB = angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; NSAID = nonsteroidal antiinflammatory drug; Tx = treatment. † Value represents the number (%) of patients. ‡ Aminoglycoside and vancomycin. § Craniotomy or craniectomy. ¶ Particularly events before development of AKI in patients with AKI.
In our study, the incidence of AKI was 10.5%, which is consistent with a previous report of 11.6%,9 although Dziedzic et al. reported an incidence of 76%.14 Zacharia et al.42 and Li et al.22 reported an incidence of AKI of 23.1% and 23%, respectively, in patients with aneurysmal subarachnoid hemorrhage and traumatic brain injury, regardless of administration of mannitol, using the RIFLE and AKIN classification systems. In the present study, data from 43 patients who did not receive mannitol were reanalyzed. Of 43 patients, 38 (88.4%) were excluded. The proportion of patients on dialysis was higher in patients not treated with mannitol than those treated with it (15.8% vs 7.8%). Five patients who did not receive mannitol treatment tended to have a lower severity of illness and did not develop AKI (data not shown). The incidence of AKI was 10.1% in 158 patients with ICH, regardless of administration of mannitol. The difference in the rate of AKI between these studies and ours may be due to different patient cohorts, comorbidities, or treatments affecting renal function. Clearly, larger studies are warranted to clarify the incidence of AKI in patients receiving mannitol infusion for increased ICP. Many studies have shown that risk for AKI increases with age.19,23 Aging kidneys with reduced renal reserve J Neurosurg / Volume 120 / June 2014
Mannitol-induced AKI TABLE 3: Univariate analysis for AKI in patients with intracranial hemorrhage* Variable
No AKI (n = 137)
AKI (n = 16)
OR (95% CI)
p Value
age ≥70 yrs male SBP ≥180 mm Hg DBP ≥110 mm Hg median Glasgow Coma Scale score medical history hypertension diabetes mellitus ischemic heart disease cerebrovascular accident ICH intracerebral hemorrhage subarachnoid hemorrhage intracerebral & subarachnoid hemorrhage subdural or epidural hemorrhage Tx during hospitalization mannitol median duration in days mean maximum dose in g/kg/day mean total dose in g/kg infusion rate ≥1.34 g/kg/day glycerol starch furosemide ACEI/ARB nephrotoxic antibiotics NSAIDs radiocontrast agent inotropic agents barbiturate decompressive craniectomy mean operative duration in hrs† laboratory data mean hemoglobin in g/dl GFR
Increased risk of acute kidney injury associated with higher infusion rate of mannitol in patients with intracranial hemorrhage Clinical article Min Young Kim, M.D.,1 Ji Hyeon Park, M.D., 2 Na Ree Kang, M.D.,1 Hye Ryoun Jang, M.D., 2 Jung Eun Lee, M.D., 2 Wooseong Huh, M.D., 2 Yoon-Goo Kim, M.D., 2 Dae Joong Kim, M.D., 2 Seung-Chyul Hong, M.D., 3 Jong-Soo Kim, M.D., 3 and Ha Young Oh, M.D. 2 Department of Internal Medicine, Seoul Medical Center; and 2Division of Nephrology, Department of Medicine, and 3Department of Neurosurgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 1
Object. Mannitol, an osmotic agent used to decrease intracranial pressure, can cause acute kidney injury (AKI). The objectives of this study were to assess the impact of mannitol on the incidence and severity of AKI and to identify risk factors and outcome for AKI in patients with intracranial hemorrhage (ICH). Methods. The authors retrospectively evaluated 153 adult patients who received mannitol infusion after ICH between January 2005 and December 2009 in the neurosurgical intensive care unit. Multivariate analysis was used to evaluate the risk factors for AKI after ICH. Based on the odds ratio, weighted scores were assigned to predictors of AKI. Results. The overall incidence of AKI among study participants was 10.5% (n = 16). Acute kidney injury occurred more frequently in patients who received mannitol infusion at a rate ≥ 1.34 g/kg/day than it did in patients who received mannitol infusion at a rate < 1.34 g/kg/day. A higher mannitol infusion rate was associated with more severe AKI. Independent risk factors for AKI were mannitol infusion rate ≥ 1.34 g/kg/day, age ≥ 70 years, diastolic blood pressure (DBP) ≥ 110 mm Hg, and glomerular filtration rate < 60 ml/min/1.73 m2. The authors developed a risk model for AKI, wherein patients with a higher risk score showed a graded association with a higher incidence of AKI. Conclusions. The incidence of AKI following mannitol infusion in patients with ICH was 10.5%. A higher mannitol infusion rate was associated with more frequent and more severe AKI. Additionally, age ≥ 70 years, DBP ≥ 110 mm Hg, and established renal dysfunction before starting mannitol therapy were associated with development of AKI. (http://thejns.org/doi/abs/10.3171/2013.12.JNS13888)
Key Words • acute kidney injury • intracranial hemorrhage • mannitol • vascular disorders
I
hemorrhage (ICH) is a severe disease with a mortality rate approaching 50% and nearly 50% of survivors left with cognitive deficits or motor dysfunction.4,10,29,34,38,41 Patients with ICH suffer from additional brain injury caused by increased intracranial pressure (ICP), which requires prompt intervention to prevent devastating neurological outcomes. ntracranial
Abbreviations used in this paper: AKI = acute kidney injury; AKIN = Acute Kidney Injury Network; AUC = area under the curve; DBP = diastolic blood pressure; GFR = glomerular filtration rate; ICH = intracranial hemorrhage; ICP = intracranial pressure; IQR = interquartile range; ROC = receiver operating characteristic; SBP = systolic blood pressure.
1340
Mannitol, a hyperosmolar agent that causes osmotic diuresis, is considered useful in the treatment of increased ICP.33 The effect of mannitol therapy is dose dependent, with higher doses providing a longer-lasting reduction in ICP.37,40 However, marked hyperosmolality may cause acute kidney injury (AKI).3,13 Previous studies have shown that extracerebral organ system dysfunctions are related to unfavorable neurological outcomes and high mortality rates in patients with brain injury and hemorrhage.16,36 In particular, AKI is associated with worse outcome in patients with brain injury.22,42 However, there is little information published regarding the incidence, risk factors, and outcome of mannitol-induced AKI. In previous case reports, AKI occurred at even small J Neurosurg / Volume 120 / June 2014
Mannitol-induced AKI (< 400 g) doses of mannitol.2,13,35,39 Pérez-Pérez et al. concluded that AKI is related to the rate of infusion rather than the total dose of mannitol.32 These results suggest that AKI may be associated with the infusion rate of mannitol rather than the dosage. The Acute Kidney Injury Network (AKIN) proposed a modified diagnostic staging system for AKI that focuses on small changes in serum creatinine.28 The AKIN classification has been validated in intensive care units and has been found to predict in-hospital mortality, the need for renal replacement therapy, and prolonged hospital stay.1 Thus, the present study was undertaken to evaluate the incidence of and associated risk factors for AKI resulting from mannitol therapy in patients treated for ICH and to determine if the infusion rate is associated with AKI, as defined by the AKIN classification.
in a relatively consistent manner. Serum electrolytes and osmolality were measured before each dose of mannitol, at least once to several times a day as needed. If serum osmolality exceeded 285 mOsm/kg, 100 ml of normal saline was infused. If serum osmolality exceeded 315 mOsm/kg, 100 ml of 0.45% normal saline was infused. The infusion rate was calculated as follows: total dose of mannitol (g)/body weight (kg)/day. A receiver operating characteristic (ROC) curve was applied to determine the cutoff mannitol infusion rate required to predict AKI. The ROC curve analysis resulted in a cutoff of 1.34 g/kg/day for AKI probability. Using this cutoff, the prediction rule showed a 75.0% sensitivity, a 65.7% specificity, and an area under the curve (AUC) of 0.71 (p = 0.007, 95% CI 0.58–0.84).
Methods
Continuous variables are expressed as the median with the interquartile range (IQR) or as the mean ± SD. Categorical variables are expressed as counts (percentages). Between-group comparisons of body weight, serum osmolality, and creatinine were made using the Student t-test or the Mann-Whitney U-test, as appropriate. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for the development of AKI in patients with ICH. Variables with a p value < 0.05 by univariate analysis were entered into multivariate analysis. Odds ratios and 95% CIs were calculated, and, based on the odds ratio, weighted scores were assigned to each of the AKI predictors. We then evaluated the incidence of AKI according to the score value. The evaluation of each AKI predictor produced an ROC curve; we used the AUC to assess the power of discrimination between patients with and without AKI. All analyses were performed using SPSS version 12.0 (SPSS, Inc.), and p values < 0.05 were considered significant.
We retrospectively studied 537 consecutive adult patients who received mannitol infusion after ICH between January 2005 and December 2009 in the neurosurgical intensive care unit. We excluded 384 patients with tumors (n = 32); renal failure requiring dialysis (n = 3); no measurement of serum creatinine before, during, and after mannitol therapy (n = 179); an administration interval of mannitol more than 2 days (n = 93); AKI (an increase in serum creatinine 0.3 mg/dl from baseline) on admission (n = 14); and patients transferred from outside hospitals during therapy (n = 63). The remaining 153 patients were included in the study. Patient data were collected from electronic medical records and databases. The demographic data and clinical data collected were age, sex, body weight, systolic blood pressure (SBP), diastolic blood pressure (DBP), and status at time of admission for all of the following conditions: diabetes mellitus, hypertension, liver disease, cerebrovascular accident, ischemic heart disease, Glasgow Coma Scale score, and type of ICH. Administration of mannitol, either alone or concurrently with other drugs (specifically furosemide, nonsteroidal antiinflammatory drugs, osmotic agents, and potentially nephrotoxic antibiotics), radiocontrast agents, inotropic agents, barbiturates, decompressive craniectomy, and adverse events such as myocardial infarction and infection were also investigated. We examined the laboratory results for serum hemoglobin, glucose, urea, creatinine, sodium, osmolality, and glomerular filtration rate (GFR). The GFR was estimated using the 4-variable Modification in Diet and Renal Disease equation.21 The study was approved by our institutional review board. The primary outcome was the occurrence of AKI, defined as a ≥ 0.3-mg/dl increase in serum creatinine level above baseline or a change of ≥ 50%, according to the AKIN classification. The severity of AKI was classified according to the AKIN system as follows: Stage 1 (increase in serum creatinine level by ≥ 50% or ≥ 0.3 mg/dl), Stage 2 (increase in serum creatinine level by > 100%), and Stage 3 (increase in serum creatinine level by > 200% or by ≥ 0.5 mg/dl with serum creatinine level of ≥ 4.0 mg/dl).28 Mannitol was administered intravenously by bolus J Neurosurg / Volume 120 / June 2014
Statistical Analysis
Results
Patient characteristics are summarized in Tables 1 and 2. The mean age of patients was 55 ± 14 years, and 77 patients (50.3%) were male (Table 1). Of the 16 patients (10.5%) in whom AKI developed, 1 (6.3%) received a total of 3 hemodialysis sessions. Complete recovery (return to baseline serum creatinine level) was observed in 11 of 16 patients, including the patient who underwent hemodialysis, a median of 4 days after hospitalization (IQR 2.5–10.5 days). The mortality rate was approximately 3.5 times higher in patients with AKI than in those without (12.5% vs 3.6%, p = 0.133). Table 3 lists the variables significantly associated with the development of AKI in patients who received mannitol in univariate analysis. The significant variables associated with AKI were age ≥ 70 years, DBP ≥ 110 mm Hg, GFR < 60 ml/min/1.73 m2, and a rate of mannitol infusion ≥ 1.34 g/kg/day in multivariate analysis (Table 4). More patients who received a mannitol infusion rate of ≥ 1.34 g/kg/day developed AKI and presented predominantly with severe AKI than those who received a lower rate (trend, p = 0.002, Fig. 1). Figure 2 shows the changes in body weight, serum 1341
M. Y. Kim et al. TABLE 1: Baseline characteristics of patients on admission* Variable
Value
no. of patients mean age in yrs no. of males mean body weight in kg mean SBP in mm Hg mean DBP in mm Hg median Glasgow Coma Scale score medical history hypertension diabetes mellitus ischemic heart disease cerebrovascular accident type of ICH intracerebral hemorrhage subarachnoid hemorrhage intracerebral & subarachnoid hemorrhage subdural or epidural hemorrhage laboratory data mean hemoglobin in g/dl mean GFR in ml/min/1.73 m2 median serum sodium in mmol/L median serum osmolality in mOsm/kg
153 55 ± 14 77 (50.3%) 62.4 ± 10.1 151 ± 33 86 ± 18 14 (9–15) 64 (42.1%) 11 (7.2%) 7 (4.6%) 16 (10.5%) 57 (36.3%) 71 (45.2%) 5 (3.2%) 20 (12.7%) 13.7 ± 1.7 95.0 ± 22.9 139 (138–141) 297 (292–304)
* Continuous variables are expressed as the median (IQR) or as the mean ± SD. Categorical variables are expressed as counts (%).
osmolality, and serum creatinine level during hospitalization in patients with and without AKI. The body weight of patients with AKI tended to increase until 2 days after stopping mannitol and, on discharge, tended to be lower in comparison with that of patients without AKI. However, the difference between groups was not statistically significant. There was a significant difference in serum osmolality and creatinine level between patients with AKI and those without AKI. A risk model was constructed using the predictors of AKI (Table 5). In the present study, risk score levels ranged between 0 and 4 points. The relationship between risk score and AKI incidence is graded: the higher the score, the stronger the association with AKI (Fig. 3). Scores were applied to predict risk for AKI after ICH. The area under the ROC curve was 0.917 (95% CI 0.851– 0.983, Fig. 4). In case of a risk score of 3, positive and negative predictive values were greater than 90% (Table 6).
Discussion
In this study, we evaluated the incidence of and risk factors for AKI in patients treated with mannitol for ICH. Of 153 patients, 16 (10.5%) developed AKI. Risk factors for AKI were identified as age ≥ 70 years, DBP ≥ 110 mm Hg, GFR < 60 ml/min/1.73 m2, and a rate of mannitol infusion ≥ 1.34 g/kg/day. The incidence of AKI increased in patients with a higher risk score. 1342
TABLE 2: Characteristics of 153 patients during hospitalization* Variable treatment mannitol† median duration of Tx in days mean maximum dose in g/kg/day mean total dose in g/kg median infusion rate in g/kg/day glycerol† starch† furosemide† ACEI/ARB† nephrotoxic antibiotics†‡ NSAIDs† radiocontrast agent† inotropics† barbiturate† decompressive craniectomy† mean operative duration in hrs§ median peak serum osmolality in mOsm/kg myocardial infarction†¶ infectious disease†¶
Value 153 (100%) 8 (5–11) 1.46 ± 0.37 9.41 ± 6.02 1.26 (1.07–1.48) 49 (32%) 101 (66%) 45 (29.4%) 25 (16.3%) 39 (25.5%) 111 (72.5%) 118 (77.1%) 40 (26.1%) 12 (7.8%) 17 (11.1%) 3.9 ± 1.1 312 (306–320) 1 (0.6%) 27 (17.6%)
* Mean values are presented ± SD; median values are presented with the IQR. ACEI/ARB = angiotensin-converting enzyme inhibitor/angiotensin receptor blocker; NSAID = nonsteroidal antiinflammatory drug; Tx = treatment. † Value represents the number (%) of patients. ‡ Aminoglycoside and vancomycin. § Craniotomy or craniectomy. ¶ Particularly events before development of AKI in patients with AKI.
In our study, the incidence of AKI was 10.5%, which is consistent with a previous report of 11.6%,9 although Dziedzic et al. reported an incidence of 76%.14 Zacharia et al.42 and Li et al.22 reported an incidence of AKI of 23.1% and 23%, respectively, in patients with aneurysmal subarachnoid hemorrhage and traumatic brain injury, regardless of administration of mannitol, using the RIFLE and AKIN classification systems. In the present study, data from 43 patients who did not receive mannitol were reanalyzed. Of 43 patients, 38 (88.4%) were excluded. The proportion of patients on dialysis was higher in patients not treated with mannitol than those treated with it (15.8% vs 7.8%). Five patients who did not receive mannitol treatment tended to have a lower severity of illness and did not develop AKI (data not shown). The incidence of AKI was 10.1% in 158 patients with ICH, regardless of administration of mannitol. The difference in the rate of AKI between these studies and ours may be due to different patient cohorts, comorbidities, or treatments affecting renal function. Clearly, larger studies are warranted to clarify the incidence of AKI in patients receiving mannitol infusion for increased ICP. Many studies have shown that risk for AKI increases with age.19,23 Aging kidneys with reduced renal reserve J Neurosurg / Volume 120 / June 2014
Mannitol-induced AKI TABLE 3: Univariate analysis for AKI in patients with intracranial hemorrhage* Variable
No AKI (n = 137)
AKI (n = 16)
OR (95% CI)
p Value
age ≥70 yrs male SBP ≥180 mm Hg DBP ≥110 mm Hg median Glasgow Coma Scale score medical history hypertension diabetes mellitus ischemic heart disease cerebrovascular accident ICH intracerebral hemorrhage subarachnoid hemorrhage intracerebral & subarachnoid hemorrhage subdural or epidural hemorrhage Tx during hospitalization mannitol median duration in days mean maximum dose in g/kg/day mean total dose in g/kg infusion rate ≥1.34 g/kg/day glycerol starch furosemide ACEI/ARB nephrotoxic antibiotics NSAIDs radiocontrast agent inotropic agents barbiturate decompressive craniectomy mean operative duration in hrs† laboratory data mean hemoglobin in g/dl GFR
