|Home | About | Journals | Submit | Contact Us | Français|
Cyclosporin A (CsA) has recently been proposed for use in the early phase after traumatic brain injury (TBI), for its ability to preserve mitochondrial integrity in experimental brain injury models, and thereby provide improved behavioral outcomes as well as significant histological protection. The aim of this prospective, randomized, double-blind, dual-center, placebo-controlled trial was to evaluate the safety, tolerability, and pharmacokinetics of a single intravenous infusion of CsA in patients with severe TBI. Fifty adult severe TBI patients were enrolled over a 22-month period. Within 12h of the injury patients received 5mg/kg of CsA infused over 24h, or placebo. Blood urea nitrogen (BUN), creatinine, hemoglobin, platelets, white blood cell count (WBC), and a hepatic panel were monitored on admission, and at 12, 24, 36, and 48h, and on days 4 and 7. Potential adverse events (AEs) were also recorded. Neurological outcome was recorded at 3 and 6 months after injury. This study revealed only transient differences in BUN levels at 24 and 48h and for WBC counts at 24h between the CsA and placebo patients. These modest differences were not clinically significant in that they did not negatively impact on patient course. Both BUN and creatinine values, markers of renal function, remained within their normal limits over the entire monitoring period. There were no significant differences in other mean laboratory values, or in the incidence of AEs at any other measured time point. Also, no significant difference was demonstrated for neurological outcome. Based on these results, we report a good safety profile of CsA infusion when given at the chosen dose of 5mg/kg, infused over 24h, during the early phase after severe head injury in humans, with the aim of neuroprotection.
Traumatic brain injury (TBI) triggers a complex cascade of pathophysiological events, leading to progressive neuronal and axonal damage (Alessandri et al., 2002; Kroemer and Reed, 2000; Lewen et al., 2001; Okonkwo et al., 1999; Povlishock and Katz, 2005; Signoretti et al., 2004). The traumatically induced influx of calcium is recognized to be a causative factor in triggering early cell death and axonal damage. Elevated intracellular calcium has been linked to an opening of the mitochondrial permeability transition pore, allowing calcium to flood the mitochondrion and causing mitochondrial swelling, the generation of oxygen free radicals, and the ultimate failure of mitochondrial function (Starkov et al., 2004; Xiong et al., 1997). Such mitochondrial dysfunction plays a significant role in TBI-induced early neuropathological events, causing the loss of ATP and increased production of reactive oxygen species (Fiskum, 2000; Lifshitz et al., 2004; Sullivan et al., 2005; Wang, 2001). These effects, in turn, lead to cell death by either necrotic or apoptotic routes, since pro-apoptotic factors such as caspase C are also released from mitochondria (Kroemer and Reed, 2000; Lewen et al., 2001; Li et al., 1997; Li et al., 2000; Sullivan et al., 2002; Uchino et al., 2002; Zou et al., 1997). Similarly, caspase 3 activation has been linked to intra-axonal cytoskeletal damage that leads to axonal failure and disconnection (Büki et al., 2000; Ferrand-Drake et al., 2003).
The complexity of TBI pathophysiology justifies the wide variety of therapeutic approaches that have been tested in experimental animal models and in humans over the last two decades, targeting numerous interrelated aspects of TBI-induced subcellular pathological mechanisms. Such therapeutic screening studies are also complicated by the fact that the patients affected by severe head injury are typically treated in intensive care units, and undergo invasive procedures such as mechanical ventilation, invasive monitoring, and multiple systemic drug therapies, and are thereby subject to multiple secondary risk factors. Accordingly, any proposed drug for use in TBI patients should first be rigorously tested for safety to avoid the possibility of further increasing morbidity and mortality in this already gravely ill patient population.
Recently, the role of cyclosporin A (CsA) as a neuroprotectant has been documented in several animal models of TBI (Alessandri et al., 2002; Okonkwo and Povlishock, 1999; Okonkwo et al., 1999, 2003; Signoretti et al., 2004; Sullivan et al., 1999, 2000b). CsA transiently inhibits the opening of the mitochondrial permeability transition pore by unbinding mitochondrial matrix cyclophillin from the pore (Broekemeier et al., 1989; Broekemeier and Pfeiffer, 1995; Szabò and Zoratti, 1991), and thus protects mitochondria in the face of increased intracellular calcium. In this fashion, CsA prevents mitochondrial destruction and the release of caspase following cellular ischemia or traumatic brain injury (Alessandri et al., 2002; Okonkwo and Povlishock, 1999; Okonkwo et al., 1999; Povlishock and Pettus, 1996; Signoretti et al., 2004; Sullivan et al., 1999, 2000 a,b). Experimental evidence suggests that CsA, via the maintenance of mitochondrial membrane potentials and bioenergetic homeostasis, may attenuate post-traumatic cytoskeletal changes and axonal injury (Okonkwo and Povlishock, 1999; Okonkwo et al., 1999; Sullivan et al., 1999, 2000a), and it may exert beneficial effects on infarct size and global ischemic damage in transient ischemia-reperfusion models (Matsumoto et al., 1999; Shiga et al., 1992; Uchino et al., 1995, 1998; Yoshimoto and Siejo, 1999). It has been proven effective in models of hypoglycemic brain damage (Ferrand-Drake et al., 2003; Friberg et al., 1998), while improving brain tissue oxygen consumption and cognitive performance after TBI (Alessandri et al., 2002). CsA's beneficial effects on infarct size and global ischemic damage have been demonstrated when the drug is given either pre-injury or post-ischemia (Friberg et al., 1998; Li et al., 2000; Yoshimoto and Siejo, 1999).
CsA also belongs to the family of calcineurin inhibitors, that act by impairing the expression of several critical cytokine genes that promote T-cell activation, resulting in a selective inhibition of the killer T-cell-mediated immune response. Its introduction in clinical practice in the 1980s resulted in a marked improvement of the outcome of solid organ transplantation; however, the use of CsA is limited in clinical practice by its adverse-effect profile, especially when the drug is administered chronically.
In the current study we designed a prospective, randomized, double-blind, dual-center, placebo-controlled trial, to evaluate the safety, tolerability, and pharmacokinetics of a single intravenous infusion of CsA in patients with severe TBI. The safety and tolerability of CsA are reported here. Other aspects of drug administration, such as its effects on cell-mediated immunological function, and brain neurochemistry and cerebral and systemic hemodynamics, have been reported previously (Mazzeo et al., 2006, 2008).
This dual-center National Institutes of Health-National Institute of Neurological Disorders and Stroke (NIH-NINDS)-funded safety study was approved by the Virginia Commonwealth University Institutional Review Board (IRB) for human research, and by the Health Center IRB of the University of Florida. Informed consent was obtained prior to enrollment from the legal representative of each patient using forms and explanations approved by the IRB.
Patients older than 16 years with severe head injury (Glasgow Coma Scale score between 3 and 8) who received a ventriculostomy and a microdialysis catheter were eligible for the study. Exclusion criteria were: both pupils fixed and dilated; blood urea nitrogen (BUN) >20mg/dL; creatinine >1.3mg/dL; history of known prior impaired hepatic or renal function; known history of cancer, pregnancy, or the inability to exclude pregnancy; current or prior use of another investigational agent within 30 days; immunosuppression; or deranged coagulation parameters preventing the placement of intracranial pressure (ICP) and microdialysis catheters.
The study was designed to enroll a total of at least 50 patients at both centers. Subjects diagnosed with a severe head injury who received a ventriculostomy and a microdialysis catheter, and for whom consent was obtained, received either an intravenous infusion of CsA or placebo within 12h of injury. The subjects were assigned to each treatment according to a randomization list in a 3:1 fashion, to allow a greater number of patients in the CsA group. The dose of CsA was 5mg/kg given over 24h as a slow continuous infusion, diluted in 250mL of 5% dextrose. The placebo was 250mL of 5% dextrose alone. This dose was decided upon based on suggested safe plasma levels from the transplant literature, and also based on neuroprotective effects seen in animal studies. CsA was purchased from Sandoz (Sandimmune®; Princeton, NJ) and dispensed, along with identical placebo infusions, by the hospital investigational drug pharmacy, who also kept the randomization codes. Blood samples for pharmacokinetic analysis were drawn at baseline, and thereafter every 12h for 6 days.
BUN, creatinine, hemoglobin, platelets, white blood cell (WBC) count, and a hepatic panel were monitored on admission, and at 12, 24, 36, and 48h, and on days 4 and 7. Potential adverse events (AEs) were also recorded, as were blood levels of CsA. Whole-blood CsA levels were measured using a commercial immunoassay kit by the hospital clinical transplant laboratory (TDx immunoassay).
On admission and before the administration of CsA, blood was collected for the determination of immunologic data: total lymphocyte count and phenotypic subsets (CD3+ mature T cells; CD4+ helper/inducer T cells; and CD8+ suppressor/cytotoxic T cells). The same analysis was repeated at 48h after admission and at day 7. Within 7 days post-injury a variety of clinical samples were collected for microbiological analysis. These included blood cultures, suctioned sputum cultures, and urine cultures. Some patients also had respiratory cultures of bronchoalveolar lavage fluid at the discretion of the surgical team. Cultures were also performed as indicated according to the clinical course during the overall period of hospital stay. These immunological data and infection complications have already been reported (Mazzeo et al., 2008).
Neurological outcome was assessed at 3 and 6 months post-injury using the 5-point Glasgow Outcome Scale (GOS), with a score ranging from 1 (dead) to 5 (good recovery) (Jennett and Bond, 1975).
A thorough graphical and exploratory data analysis was carried out to aid in the interpretation of the mean differences, and to perform diagnostic operations. A within-patient trend analysis was performed to diagnose multivariate outliers.
A mean difference test was used to test the differences between the treatment and placebo groups in each indicator. We first performed a diagnostic test for the homogeneity of the variances (Bartlett test), and if the variances were equivalent, a standard t-test was used to evaluate the mean difference. If the variances were not equivalent, a corrected t-test was employed (Welch test).
A multivariate repeated-measures linear model with covariates was used to test the stability of the measures over time, especially linear trends (increasing or decreasing values). The significance of these results were evaluated with the multivariate F-test statistic.
For the analysis of neurological outcome two tests were performed: the Pearson test of independence and the Fisher exact test for the count table.
Fifty adult severe TBI patients (41 males and 9 females) were enrolled in this study over a period of 22 months, starting in January 2003. Thirty-seven patients were randomly assigned to receive 5mg/kg of CsA within 12h of injury, which was diluted in 250mL of 5% dextrose and infused over 24h. Identical placebo infusions without the CsA were given to 13 patients.
Age, Glasgow Coma Scale score (GCS) on admission, and Injury Severity Score (ISS) of the two groups are presented in Table 1. Of the total of 50 patients, 37 were enrolled at the Neuroscience Intensive Care Unit of the Virginia Commonwealth University Medical Center in Richmond, and 13 patients were enrolled at the Department of Neurosurgery of the University of Florida in Gainesville.
Results from patient 41, randomized to receive CsA, were excluded from further safety analysis due to his multiple severe traumas with hemodynamic instability, and life-threatening multiple organ dysfunction that interfered with the analysis of safety of CsA. All his parameters were outliers and therefore he was not similar to the other patients at baseline, so he was excluded from the analysis. Thus in this study we present safety data from 49 patients, 36 receiving CsA and 13 receiving placebo.
This patient was a 52-year-old man admitted after a multiple rollover high-speed sports-car crash requiring prolonged extrication. GCS on admission was 4, Marshall CT classification was 3, and ISS was 45. The head CT scan on admission showed areas of subarachnoid hemorrhage and cortical contusion, a small epidural hemorrhage at the anterior aspect of the right middle fossa, and subdural hemorrhage along the tentorium and the falx, suggestive of diffuse axonal injury. Left frontal, right temporal, and skull base fractures were present, and multiple maxillofacial injuries were also documented.
He also sustained mesenteric artery injury causing ischemic bowel. On the first day of admission the patient therefore underwent exploratory laparotomy, with ligation of bleeding mesenteric artery vessels and resection of ischemic bowel. The intraoperative period was characterized by severe hypotension and he received multiple transfusions during the clinical course (>35 units of red blood, platelets, and plasma). During his hospital stay he underwent cardiocirculatory arrest requiring cardiopulmonary resuscitation, and placement of a chest tube for left pneumothorax. He also underwent placement of an inferior vena cava filter. Several days later, acute renal failure likely secondary to ischemia/hypotension-induced acute tubular necrosis developed, and he underwent continuous veno-venous hemofiltration. On day 19 his family withdrew further care due to poor prognosis.
A significant difference was seen for BUN at 24 and 48h between CsA and placebo patients: 10.1±3.4 versus 7.8±3.5mg/dL (at 24h; p=0.048) and 9.8±3.8 versus 7±2.7mg/dL (at 48h; p=0.017) in CsA and placebo patients, respectively (Fig. 1). However, mean BUN values remained within their normal range (6–22mg/dL) in both groups throughout the study period, although a trend of increasing BUN values was observed in both groups at day 7, with mean BUN values of 14.8±4.8 and 17.4±4.5mg/dL in the CsA and placebo patients, respectively.
Multivariate analysis focusing on BUN values within the CsA group showed there was no significant change over time.
There were no significant differences in creatinine values between CsA- and placebo-treated patients during drug administration and over the monitoring period (Fig. 2). Overall, creatinine remained within the normal range (0.7–1.4mg/dL) in both groups.
Multivariate analysis of creatinine values in the CsA group showed there were no significant changes over time. The mean baseline creatinine value in the CsA-treated group was 0.9±0.2mg/dL, and it remained stable at 0.8±0.2 at 48h, and at 0.75±0.2 and 0.7±0.2 at days 4 and 8, respectively.
Mean AST values showed no significant difference between the placebo- and CsA-treated groups, and were lower in the CsA-treated group than in the placebo-treated group throughout the monitoring period. At 48h, mean AST levels (normal range: 0–75U/L) were 50.3±38.4 and 63.5±51.5U/L in CsA-treated and placebo-treated patients, respectively. Toward the end of the first week, all liver function enzyme test values trended upward and exceeded the normal range, but in all patients this later normalized, and no clinical liver abnormalities developed.
The same trend described for AST applied for ALT, with mean ALT values showing no significant difference between the placebo-treated and CsA-treated groups, and levels remaining lower in the CsA-treated group than in the placebo-treated group throughout the monitoring period. At 48h, mean ALT levels were 42.4±25.5 and 49.6±39U/L, in CsA-treated and placebo-treated patients, respectively. The normal range for ALT is 0–50U/L.
Our study demonstrated no significant difference in total bilirubin levels between the two groups throughout the monitoring period. Total bilirubin remained within the normal range (0.2–1.3mg/dL) in both study groups.
No significant difference in total alkaline phosphatase levels was seen between the two groups throughout the monitoring period. At 48h, mean alkaline phosphatase levels (normal range: 0–126U/L) were 78±20 and 77±17U/L, in CsA-treated and placebo-treated patients, respectively. As with AST and ALT, an increase in alkaline phosphatase was seen at days 4 and 8 in both groups, but was only a little above the normal range, and no overt hepatic clinical changes were seen; values normalized after 2–3 weeks.
A significant difference was seen in WBC counts only at 24 hours in CsA-treated versus placebo-treated patients: 15.7±4.8 versus 12.5±3.2×109/L (p=0.02), respectively (Fig. 3). Baseline WBC counts prior to CsA or placebo administration were 17.2±6.3 and 15.4±4.7×109/L in the CsA and placebo patient groups, and at day 7 the values were 12.7±4.1 and 14.5±5.1×109/L, respectively. Thus the values were above the normal range (3.2–11×109/L) throughout the measurement period, consistent with the effect of a severe trauma-induced stress response or infection.
Throughout the study period no significant differences were demonstrated between the CsA and placebo groups for hemoglobin. Mean baseline hemoglobin was 12.1±2.3 and 12.8±1.9g/dL in CsA-treated and placebo-treated groups, respectively. Hemoglobin levels decreased to 10.3±1.8 and 10.4±2.1g/dL at 24h, and then remained stable with values of 10.2±1.2 and 9.9±1 at day 7, in CsA-treated and placebo-treated patients, respectively.
Our study demonstrated no significant difference in platelet count between the two study groups throughout the monitoring period, with values remaining within the normal range (normal: 165–350×109/L). Baseline platelet counts were 235±105 and 225±68×109/L, and at day 7 were 318±136 and 326±169×109/L, in CsA-treated and placebo-treated patients, respectively.
The occurrence of adverse events (AEs) in the CsA- and placebo-treated patients during the first week following drug infusion were recorded, and the analysis demonstrated no significant difference between the two groups. An AE was defined as any clinical event that was treated or found to be clinically significant by the principal investigator. Numbers and percentages of AEs in the CsA-treated and placebo-treated groups, respectively, were as follows.
The effect of CsA infusion on blood pressure after head injury has been previously reported (Mazzeo et al., 2008). After the initiation of the CsA infusion, an increasing mean arterial pressure (MAP) was seen in the drug treated group, with MAP levels reaching a mean value of 105mm Hg at 56h after injury (Fig. 4A). MAP values were then maintained at a higher level in the CsA-treated group than the placebo-treated group throughout the monitoring period. The difference in MAP between the two groups was statistically significant at p<0.01 at multiple time points; however, MAP never exceeded the physiologic range. Significantly increased cerebral perfusion pressures (CPPs) became apparent in the CsA-treated patients at 68h after injury, with mean values of 90mm Hg in the CsA-treated group and 81mm Hg in the placebo-treated group at this time point (Fig. 4B). The recorded CPP values were subsequently significantly higher in the CsA-treated group than in the placebo-treated group throughout the 7-day monitoring period (p<0.01) (Mazzeo et al., 2008). Nevertheless, no significant difference was recorded in the two groups in the occurrence of clinically significant hypertension, confirming that even if MAP was maintained at higher levels in CsA-treated patients, these values remained within the normal range, and did not require therapeutic intervention to avoid clinical adverse effects.
No significant difference was demonstrated in the incidence of infections or sepsis between the two study groups. These findings mirror those reported in a previous article, in which we clearly demonstrated that no significant difference exists in the studied immunologic parameters in the placebo- versus CsA-treated groups at any time point (Mazzeo et al., 2006). Furthermore, we could not demonstrate any other significant covariable factors in the incidence of lung infections after admission, other than the early impairment of T-cell activation secondary to injury that was not due to the administration of CsA (Mazzeo et al., 2006).
Four of the 37CsA-treated patients showed high CsA blood levels (>1000ng/mL). In Table 2, clinical data for these patients are separately reported. Their CsA blood levels throughout the monitoring period are shown in Figure 5A. No significant effect on any safety parameters was seen in this subgroup with blood levels transiently >1000ng/mL. Mean blood CsA levels for the entire CsA-treated group are shown in Figure 5B.
In Table 3 is presented a summary of the clinical course of the 11 patient deaths that occurred during this clinical trial, along with type and severity of injury, clinical presentation on admission, and cause of death.
Neurological outcome was assessed at 3 and 6 months post-injury using the Glasgow Outcome Scale (Jennett and Bond, 1975). We had three missing scores in the CsA group at 6 months due to loss to follow-up, so the analysis for outcome was performed on a total of 47 patients (13 placebo patients and 34CsA patients; Fig. 6). The distribution of outcome in the 5 GOS categories at 6 months in the placebo and CsA groups, respectively, were: good (4 and 8), moderate (4 and 6), severe (3 and 7), vegetative (0 and 4), and death (2 and 9). For analytical purposes the outcomes were dichotomized as bad outcome (dead, vegetative, or severe disability) and good outcome (moderate disability or good recovery).
Fisher's exact test demonstrated that the difference between the two groups was not significant both at 3 months (p=0.7) and at 6 months (p=0.3). The odds ratios adjusted for age and GOS scores also showed no statistically significant differences.
This study demonstrated only transient, yet statistically significant, differences for BUN at 24 and 48h, and for WBC at 24h, between the CsA-treated and placebo-treated patients. These modest differences were not clinically significant, in that they did not negatively impact patient course. Both BUN and creatinine values, markers of renal function, remained within normal limits over the entire monitoring period in both placebo-treated and CsA-treated patients. There were no significant differences in any other mean laboratory values, or in the incidence of AEs at any other measured time point. Also, no significant differences were demonstrated for neurological outcome. These results demonstrate the good safety profile of CsA infusion when given at the chosen dose of 5mg/kg infused over 24h during the early period after severe head injury in humans.
CsA is a small polypeptide with a molecular weight of 1.203kDa, that was introduced into clinical practice in 1983 for rejection prophylaxis in transplant patients, and it acts through selective inhibition of the immune response. CsA exerts its immunosuppressant action by binding to an intracellular protein, cyclophillin, and the CsA-cyclophillin complex then inhibits the activity of calcineurin. As a calcineurin inhibitor, CsA selectively targets interleukin-2 (IL-2)-dependent T-cell proliferation, thus blocking the cell-mediated killer T-cell immune response (Danovitch, 2005). Calcineurin inhibitors are the mainstay of prevention of allograft rejection in modern transplant practice, and are approved for this indication by the U.S. Food and Drug Administration (Kapturczak et al., 2004).
With the increased clinical use of CsA in transplantation medicine and to treat autoimmune disease, significant CsA-related side effects have become apparent in the organ transplant population, and the regulation of immunosuppression remains a challenge in the care of these transplant patients (Shaw et al., 1996). Furthermore, CsA is a drug with a narrow therapeutic index, and is characterized by significant pharmacokinetic inter- and intra-individual variability and unpredictability in its pharmacodynamic effects, mandating strict blood level monitoring and careful titration (Danovitch, 2005; Kapturczak et al., 2004, Paul and de Fijter, 2004; Ptachcinski et al., 1986; Schiff et al., 2007).
Several important side effects of CsA have been described in chronically treated transplant patients, in particular nephrotoxicity, acute microvascular disease, electrolyte abnormalities, and hypertension. There are also reports of neurotoxic manifestations, such as tremor, dysesthesia, and seizures, and gastrointestinal symptoms such as dyspepsia, nausea, vomiting, hypertrichosis, hyperlipidemia, glucose intolerance, and hyperuricemia, as well as an increased incidence of thromboembolic events (Danovitch, 2005; Miller, 1996; Min and Monaco, 1991; Remuzzi and Perico, 1995). Among these, CsA-induced acute renal dysfunction is by far the most significant and limiting adverse effect, manifest as reversible increases in BUN and creatinine levels (Al Aly, 2005; Cattaneo et al., 2004; Fellstrom, 2004; Paul and de Fijter, 2004; Rezzani, 2004; Shaw et al., 1996).
The aim of this study was to evaluate the safety profile of cyclosporin A when given in the early phase post-TBI for its presumed neuroprotective effects. Recently, Piot and associates reported no significant adverse effects of CsA when administered in humans at a dose of 2.5mg/kg for the purpose of attenuating myocardial reperfusion injury after myocardial infarction (Piot et al., 2008). Furthermore, the results of a randomized, placebo-controlled dose-escalation trial investigating the role of CsA when administered within 8h of injury to a population with severe TBI have been recently published, and showed no difference in mortality or other adverse events in patients receiving CsA versus those receiving placebo (Hatton et al., 2008).
When a drug is introduced into clinical use for a new indication, its proposed benefits must be critically evaluated against its potential adverse effects. This is especially true for those with acute trauma, who are critically ill and often physiologically unstable, and must undergo multiple invasive procedures while receiving multiple drug therapies. These patients are especially prone to develop pneumonia and multiple organ dysfunction, especially if they remain in the ICU for prolonged periods. The incidence is proportional to the severity of the primary injury, and is influenced by the presence of pre-existing disease, such as diabetes and ischemic heart disease. Superimposed infection frequently develops, as well as metabolic derangements, and both contribute to increased morbidity and mortality in these critically ill patients. Therapeutic management in the ICU should focus on preventive measures, with the aim of minimizing organ-specific dysfunction, thus improving outcome.
In TBI patients receiving CsA, the possibility of interactions with other drug therapies should be considered and requires careful vigilance (Danovitch, 2005; Kapturczak et al., 2004). Particularly problematic are drugs that may decrease CsA concentrations by induction of cytochrome P450 activity, such as anticonvulsants (e.g., barbiturates, phenytoin, and carbamazepine), and drugs that may increase CsA concentrations via inhibition of cytochrome P450 or by competition for its pathways, such as calcium channel blockers, antifungal agents (especially ketoconazole), histamine blockers, or macrolide antibiotics. The administration of any additional potentially nephrotoxic drugs such as aminoglycosides, nonsteroidal anti-inflammatory drugs, or angiotensin-converting enzyme inhibitors, should be carefully considered and avoided whenever possible (Danovitch, 2005).
In this study, the possible toxic effects of CsA were assessed in detail during the drug infusion period and over the first week after admission in a large number of severely head-injured patients admitted to the ICU, and no clinically significant AEs due to drug interactions were found.
In relation to the potential nephrotoxic effect of CsA in our patient population, we demonstrated a significant difference in mean BUN concentration at 24 and 48h between CsA- and placebo-treated patients. However, these differences were not clinically relevant since these values remained within their normal limits.
As observed in transplant patients with long-term high-dose CsA treatment, acute and chronic nephrotoxicity are common (Al Aly et al., 2005; Benigni et al., 1999; Paul and de Fijter, 2004; Shaw et al., 1996). Acute nephrotoxicity is associated with reduced renal blood flow, reduced glomerular filtration rate, and increased renal vascular resistance (Fellstrom, 2004). CsA administration thus produces a highly dose-dependent and rapidly reversible renal vasoconstriction that particularly affects the afferent arteriole and leads to renal hypoperfusion (Cattaneo et al., 2004; Danovitch, 2005; Fellstrom, 2004). This acute nephrotoxicity is thought to be caused by a local imbalance of vasoconstrictor and vasodilator mediators, such as increased endothelin activity, increased levels of angiotensin II, and reduced availability of vasodilator prostaglandins and nitric oxide, as well by activation of the renal sympathetic nerves (Cattaneo et al., 2004; Danovitch, 2005; Fellstrom, 2004; Paul and de Fijter, 2004). Most importantly, this effect is reversible with CsA dose reduction or withdrawal, and is not followed by renal histological changes. It is seldom seen in those with less than 2–3 months of dosing. Conversely, chronic CsA nephrotoxicity represents a more serious concern in patients receiving long-term (lifetime) CsA treatment, and it is associated with progressive and irreversible renal interstitial fibrosis, and clinically significant renal dysfunction eventually leading to end-stage renal failure (Cattaneo et al., 2004; Danovitch, 2005; Fellstrom, 2004; Paul and de Fijter, 2004). Renal exposure to and accumulation of systemically derived cysteinyl leukotrienes, such as leukotriene C4 and D4 products, may be underlying factors in CsA nephrotoxicity (Aleo et al., 2008). Our study protocol consisted of brief administration (24h) of a therapeutic dose of CsA within the recognized dosing range of 5–6mg/kg, and was associated with none of these effects.
No significant difference in mean serum creatinine level between the CsA and placebo groups was shown in our study, confirming the safe renal profile of CsA when given as a putative neuroprotective drug in the early phase after severe human TBI.
The results of this study demonstrated no significant difference between CsA- and placebo-treated patients in the evaluation of parameters of hepatic function, such as AST, ALT, bilirubin, and alkaline phosphatase levels. Increased mean values for AST, ALT, and ALP, above the normal range, were however recorded on days 4 and 8 in both treatment groups, but no clinical signs of liver dysfunction were seen, and in all cases this effect disappeared in a few weeks.
In the literature there are reports of similarly transient gastrointestinal effects of chronic CsA administration, such as subclinical mild and self-limiting dose-dependent elevations in serum aminotransferase levels and mild hyperbilirubinemia in nearly half of the treated patients. This did not seem to be CsA-related, as this was also seen in the placebo group. In the literature, hyperbilirubinemia has been related to disturbed bile secretion, rather than hepatocellular damage, and no hepatic histological lesions have been described in humans after the use of CsA (Aleo et al., 2008; Danovitch, 2005; Min and Monaco, 1991; Rezzani, 2004).
In this study we found a transient significant difference in WBC counts at 24h. No significant differences were detected in hemoglobin concentrations and platelet counts between the two study groups during the monitoring period. These results support the good hematological safety profile of CsA when given at 5mg/kg over 24 hours.
Many factors may influence WBC counts, hemoglobin levels, and platelet counts in TBI patients during their ICU stay, such as age, concomitant disease, risk of infection, multiple drug therapies, colloid or crystalloid replacement therapy, and length of hospital stay.
Trauma is a multisystemic disease that interferes with the function of multiple organs and systems, even those not directly related to the site of the initial impact. Infectious complications and sepsis have been reported in up to 50–70% of severe TBI patients, and worsen clinical outcomes, increasing both late morbidity and mortality (Helling et al., 1988; Hoyt et al., 1990; Quattrocchi et al., 1991a,b). Furthermore, severely head-injured patients have other factors known to increase infection rates, such as a prolonged need for ventilatory support, prolonged sedation and paralysis, aspiration of gastric contents, and invasive monitoring.
It is also well-known that immunological function may be impaired in the early phase after TBI, as well in as trauma in general, due to a complex multifactorial effect of suppression of T-cell mediated immune function, and changes in soluble mediators, such as an increase in PGE2 biosynthesis, and reductions in the levels of the key immunoregulatory cytokines IL-2 and IFN-γ (Boddie et al., 2003; Quattrocchi et al., 1990, 1991a,b, 1992). The administration of any medication that might induce further immunological suppression should therefore be undertaken with caution, in order to avoid any additional or synergistic suppression of immune function after severe TBI. We have previously reported that there were no significant differences between the placebo- and CsA-treated patients in this clinical trial with regard to immunological parameters, such as total lymphocyte count and phenotypic cell subsets (CD3+ mature T cells, CD4+ helper/inducer T cells, and CD8+ suppressor/cytotoxic T cells), or the incidence of infection (Mazzeo et al., 2006). No significant difference was seen in the incidence of infection or sepsis between the two study groups detailed here.
In the literature there are reports of CsA-induced hypertension, which is a potentially serious side effect in patients receiving the drug as an immunosuppressive agent, due to stimulation of the renin-angiotensin system, as well as sympathetic activation, endothelin release, and nitric oxide production, all of which play important roles in the pathogenesis of hypertension (Gardiner et al., 2004; Roullet et al., 1994; Shaltout and Abdel-Rahman, 2003).
The results of our study strongly suggest a significant vasoactive effect of CsA, and we previously reported a significant transient hypertensive effect in the CsA-treated group (Mazzeo et al., 2008). This may be beneficial in the early phase after TBI, especially since maintenance of CPP is an important goal of TBI treatment, and in our trial the need for vasopressors was reduced in the CsA-treated group (Mazzeo et al., 2008).
A spectrum of neurologic complications has been described in patients receiving CsA as an immunosuppressant. Clinical manifestations of CsA-associated neurotoxicity include tremors, restlessness, dysesthesias, headaches, and seizures (Miller, 1996; Danovitch, 2005). In our trial neurotoxicity was not evident in the CsA-treated group; the incidence of seizures was 5% in the CsA-treated group and 8% in those receiving placebo.
The blood CsA levels are shown in Figure 5. As can be seen in Figure 5B, the highest mean blood levels for the entire CsA group was ~ 500ng/mL, and these levels remained high for about 12h after the infusion ended. In four patients the blood level exceeded 1000ng/mL, and this was about 2–3 times higher than the levels seen in the group as a whole. The reasons for these higher blood levels were not clear, but no adverse effects were seen in any of the monitored safety parameters in this small patient group (Table 2), although one of these four patients died.
Our results demonstrate that there was no significant difference in neurological outcomes at 3 and 6 months in the two study groups. As shown in Table 3, there was a higher mortality rate in the patients receiving CsA. This appears to be due to the severe systemic sequelae suffered by the CsA patient group, rather than any possible drug effect. In fact, as shown in Table 1, the median ISS scores were higher in CsA-treated patients than in those receiving placebo (34 versus 25), as was the percentage of patients with critical ISS scores (89% versus 67%), suggesting that the CsA-treated patients were more severely injured than those receiving placebo. Data analysis also demonstrated that 8 patients in the CsA-treated group showed improvements in GOS scores over time.
This study demonstrates the good safety and tolerability profile of CsA when it is administered early after severe TBI with the goal of neuroprotection. Only brief, transient significant differences were seen for BUN at 24 and 48h, and for WBC counts at 24h, between the CsA- and placebo-treated groups; these differences were small and were not clinically significant. There were also no significant differences in the other studied clinical parameters, or in the incidence of adverse events, at any time point. No significant differences were seen with regard to neurological outcomes in the two study groups.
Pharmacokinetic studies have demonstrated that CsA administration is complicated by wide variations in inter- and intra-patient metabolic levels (primarily mediated by changes in cytochrome P450 metabolism), and thus drug levels must be carefully monitored during treatment.
A large Phase III efficacy trial of CsA in severe TBI is currently under peer review by the NIH-NINDS Clinical Trials study section, and if approved, it will be performed in about 50 centers throughout the U.S. to further assess the efficacy of this agent for this indication.
This study was funded by the NIH-NINDS as part of project grant no. NS12587 to M.R.B. (the primary investigator), and by the Lind-Lawrence Foundation and the Reynolds Foundation.
No competing financial interests exist.