Abstract

Background

The early postoperative period is a critical time for patients after receiving a decompressive craniotomy. Prompt detection and early management of postoperative recurrent/residual hemorrhagic complications may dramatically improve clinical outcomes.

Methods

The present cohort retrospective study involved 135 patients who received decompressive craniotomy and intensive care unit (ICU) supervision as life-saving measures. The purpose of the study was to evaluate the effects of propofol sedation on the clinical outcome during the ICU stay. The patients' demographic data, hemodynamic variables, the dose of propofol used during the first 48 hours after surgery, residual/recurrent blood clot volume after surgery, and neurologic and clinical outcomes were reviewed. The propofol dosages used for sedation were further divided into three categories: < 0.66, 0.66-3.33 and > 3.33 mg/kg/hr, based on the doses infused during the first 12 hours after surgery.

Results

Our results indicated that the patients of the propofol-sedated group had a significantly smaller amount of residual/recurrent blood clot (p < 0.05) than did those of the non-sedated group. The 30-day mortality rate was significantly lower in the propofol-sedated group (p < 0.05) than in the non-sedated group. Among the propofol-sedated patients, those who received a dose of 0.66-3.33 mg/kg/hr in the first 12 hours after surgery achieved significantly improved clinical and neurologic outcomes than those who received either more than 3.33 mg/kg/hr or less than 0.66 mg/kg/hr of propofol.

Conclusion

Our results support the use of propofol sedation during the early postoperative period after craniotomy in hemorrhagic stroke patients, because it improved both neurologic and clinical outcomes. However, early postoperative use of propofol sedation at larger dosages warrants special attention.

Keywords

craniotomy; intensive care units; morbidity; propofol;

1. Introduction

Postoperative residual/recurrent hematoma is one of the complications that occur most frequently in patients undergoing a decompressive craniotomy for hemorrhagic stroke. The early postoperative period is,  therefore,  a  critical  time  for  these  patients,  because  prompt  detection  and  early  management  of  postoperative  recurrent/residual  hemorrhagic  complications may dramatically improve their clin-ical outcomes. Specifically, intraoperative and early postoperative hypertension is a well known factor leading  to  postoperative  recurrent  intracranial  hemorrhage.¹⁻³  Basali  et  al  showed  that  in  post-craniotomy patients who experienced episodes of intraoperative  and  early  postoperative  hyperten-sion, the incidence of recurrent intracranial hem-orrhage was up to 62%. In contrast, those patients who did not have early postoperative hypertension had a lower incidence of recurrent intracranial he-morrhage  of  34%.¹  The  Society  of  Critical  Care  Medicine, therefore, has provided valuable guide-lines on the sustained use of sedatives and analge-sic  agents  in  surgically  treated  and  critically  ill  adults.^3,4 For patients who underwent intracranial surgery,  several  clinical  studies  have  also  shown  the  importance  and  potential  benefits  of  postop-erative sedation.⁴⁻⁶

Sedatives are given to patients in the intensive care unit (ICU) to provide an amnesic, hypnotic, and pain-free condition, as well as to relieve agitation or anxiety.^7,8 Propofol, a short-acting γ-aminobutyric acid  agonist,  provides  easily  controlled  sedation.  Many of its pharmacological properties are useful for patients who have had neurosurgical surgery.⁵ Propofol provides a beneficial effect on compromised cerebral  hemodynamics,  in  addition  to  its  specific  potential  as  an  effective  neuroprotectant  for  the  damaged brain. The agent also has favorable phar-macokinetics  and  a  high  quality  recovery  profile,  even  under  conditions  of  prolonged  duration  of  infusion.^3,5,9 Therefore, we acknowledge and pro-pose the use of propofol sedation because it might decrease the incidence of early postoperative hy-pertension  and,  thereby,  decrease  the  incidence  and extent of recurrent hematoma following crani-otomy. Accordingly, we conducted a retrospective cohort study to evaluate the influence of propofol application during the ICU stay on the clinical out-come in hemorrhagic stroke patients who had un-dergone decompressive craniotomy.

2. Materials and Methods

Patients admitted to the University Medical Center for  craniotomy  because  of  cerebral  hemorrhage between 2000 and 2005 were included. Clinical de-mographic data including age, sex, clinical symptoms, Glasgow  Coma  Scale  (GCS)  scores  at  the  time  of  admission, postoperative care and discharge, diag-nosis, systemic disease, complications, length of stay in the ICU, and therapeutic outcome up to 6 months after surgery were collected. In addition, hemody-namic and intracranial pressure (ICP) variables, drugs used during postoperative ICU care, blood clot vol-ume before and after surgical decompression, and the dose of propofol used for sedation were indi-vidually recorded.

An intracranial blood clot was diagnosed by com-puted tomography (CT) performed 4−8 hours after surgery pursuant to the standardized institutional protocol stipulating a slice thickness of 4 mm for the posterior fossa and 8 mm for the supratentorial region with  the  slice  spacing  being  equal  to  slice  thick-ness. The  hematoma  location  was  further  catego-rized into one of four sites: (1) putaminal; (2) thalamic and  intraventricular;  (3)  lobar;  (4)  cerebellar. Clot volume  was  calculated  by  computer-assisted plani-metric analysis using the Image Pro plus 5.1 software package (Media Cybernetics, Silver Spring, MD, USA). Briefly, the hematoma was traced and measured. The  calculated  hemorrhagic  areas  at  the  predeter-mined coronal sections were then compiled to ob-tain the clot volumes in cubic centimeters (cm³).

The  need  for  the  use  of  sedatives,  including  propofol, was subjectively judged by neurosurgeons according to the clinical condition of the craniot-omized patients. To distinguish the effects of pro-pofol, we divided patients into a propofol-sedated group and a non propofol-sedated group. The esti-mate of the dosage of propofol was based on the total dose infused during the first 12 hours after sur-gery, given according to the clinical judgment of the neurosurgeon, which could sufficiently pacify anxi-ety and agitation, while at the same time maintain stable hemodynamics. Thus, the dosage was graded as mild (< 0.66 mg/kg/hr),  moderate  (0.66−3.33 mg/kg/hr) or high (> 3.33 mg/kg/hr).

Data are expressed as mean ± standard  deviation. All statistical analyses were performed using SPSS software (SPSS Inc., Chicago, IL, USA). Quantitative variables  were  compared  using  one  way  ANOVA  if  they were normally distributed, or the Mann-Whitney U test if they were not. Qualitative variables were compared using the χ² test or Fisher’s exact test. A p value of less than 0.05 was considered to be sta-tistically significant.

3. Results

A total of 135 patients who had received an emer-gency  decompressive  craniotomy  for  intracranial clot removal at the University Medical Center were included  in  the  study.  The  clot  location  in  the  putaminal areas occurred in 80 patients (59.3%), in the putamino-thalamic nuclei in 25 patients (18.5%), in  the  cerebellum  in  17  patients  (12.6%),  and  in  subcortical (lobar) areas in 13 patients (9.6%).

We found that the age of patients of the propofol-sedated group was significantly less than the non propofol-sedated patients; therefore, age may be an important factor determining the need for sed-ative use during the early postoperative period and affect patient outcome (Table 1). Compared with the  non  propofol-sedated  patients,  the  propofol-sedated patients had a superior GCS score at the time of discharge (p < 0.05; Table 1 and Figure 1). At the time of follow-up brain CT images taken within the first  48  hours  after  surgery,  the  propofol-sedated patients  also  had  a  significantly  smaller  size  of  recurrent/residual  intracranial  blood  clot,  com-pared with the patients without propofol sedation (p < 0.05; Table 1). Additionally, the 30-day mortality rate was significantly lower in the propofol sedated group  compared  with  the  non  propofol-treated  group (p < 0.05; Table 1). Other hemodynamic pa-rameters, including arterial blood pressure and ICP, did not differ significantly between the propofol-sedated and non propofol-sedated patients (Table 1 and Figure 2).

Among  the  propofol-sedated  patients,  those  who received propofol at a dose of 0.66−3.33 mg/kg/hr had a significantly higher GCS at discharge than the other two dosage groups (p < 0.05;  Table 2).  This  improved  neurologic  outcome  could  be  attributed  to  a  decrease  in  residual/recurrent  clot volumes after surgery (p < 0.05;  Table  2),  but might  not  be  related  to  a  change  in  ICP,  body  temperature, or systemic arterial blood pressure (p > 0.05, respectively; data not shown). However, those  who  received  propofol  sedation  at  a  dose  >  3.33  mg/kg/hr  for  the  first  12  hours  after  sur-gery had a significantly larger recurrent blood clot within  the  brain  after  craniotomy  for  hematoma  evacuation (p < 0.05) than did the other two groups of patients.

Download full-size image

Download full-size image

Figure 1 Comparison of the Glasgow Coma Scale (GCS) scores of patients during the various stages. Data are presented as mean ± standard deviation. *p < 0.05, vs. non propofol-sedated group.

Download full-size image

Figure 2 (A) Variation in intracranial pressure (ICP) between the two groups during the 12 hours in the intensive care unit (ICU). (B) Postoperative (Postop) mean arterial blood pressure (MAP) at the beginning of 12 hours in the ICU. Data are presented as mean ± standard deviation.

Download full-size image

4. Discussion

This retrospective study demonstrated the benefi-cial effect of propofol sedation during the early post-operative  period  after  decompressive  craniotomy  for hemorrhagic stroke. We suggest that the effect is statistically associated with an improvement in both  neurological  and  clinical  outcomes  partly  as  a consequence of decreasing the residual and re-current intracranial clot. The propofol-induced re-duction  in  recurrent  intracranial  clot  volume  and  improvements  in  clinical  outcomes  could  not  be  attributed to the changes in hemodynamics, prin-cipally  blood  pressure  or  ICP,  since  they  did  not  significantly differ between propofol-sedated and non-sedated patients.

Prompt  detection  and  management  of  hemor-rhagic complications, which most commonly occur at  the  early  postoperative  stage,  are  crucial  for  the care of neurosurgical patients. The control of postoperative  hypertension  is  essential  in  neuro-surgical  patients  because  more  than  90%  of  the  patients in the postanesthesia recovery room had a blood pressure exceeding the preoperative value by  20%.⁹⁻¹¹  Hypertension,  therefore,  is  a  well-known factor leading to an increased risk of post-operative  intracranial  hemorrhage  and  cerebral hyperemia.^1,12 Moreover, acute hypertension after craniotomy  may  increase  morbidity  and  mortality  by  exacerbating  cerebral  edema,  raising  ICP,  or  disrupting  the  delicate  postop  erative  hemostatic  state.^2,13,14 The frequency of postoperative hyper-tension has been reported to be 6% after general surgery   and   35−50%   after   cardiac   surgery.^15,16

Basali et al reported that 62% of patients with in-tracerebral  hemorrhage  had  hypertension  in  the  initial 12 hours postoperatively as opposed to 25% of control patients in their study.¹ It has also been reported  that  47%  of  intracranial  tumor  surgery  patients suffered one or more complications in the first 4 hours postoperatively.^3,17,18 How ever, there are to date no level I−II data available addressing the beneficial effects of postoperative sedation on patient  outcome  following  craniotomy.  From  our  retrospective data, we support the use of propofol sedation  following  craniotomy  for  hemorrhagic  stroke  because  it  can  improve  GCS  score  and  de-crease postoperative 30-day mortality rate.

Kelly  et  al  compared  propofol/morphine  seda-tion  with  morphine  sedation  alone  in  intubated  head-injured patients.¹⁹ They found that the pro-pofol/morphine sedation group had a lower ICP on the  third  day,  but  a  similar  long-term  outcome,  compared with the morphine group. It was there-fore  suggested  that  propofol-based  sedation  may  be a superior regimen to opiate-based sedation in intubated head-injured patients.²⁰ Chiu et al also reported that propofol offered improvement in the recovery phase in patients with head injury by de-creasing  the  rise  in  ICP  and  by  maintaining  ade-quate cerebral perfusion pressure.²¹ In this study, however,  we  failed  to  detect  notable  hemody-namic  or  ICP  stabilization  effects  offered  by  pro-pofol sedation at the optimal dosage (0.66−3.33 mg/kg/hr), although we could not completely exclude the  fact  that  the  use  of  propofol  at  this  dosage  range might have transiently stabilized the hemo-dynamics, thereby decreasing the formation of re-current  intracranial  clots.  Curiously,  although  the  low-dose  propofol  (< 0.66 mg/kg/hr)  sedated  pa-tients  had  more  favorable  hemodynamics  (blood  pressure) and a comparable size of recurrent blood clots, compared with those patients with propofol sedation   at   the   optimal   dosage   range   (0.66−3.33 mg/kg/hr), the former indeed had a relatively unfavorable  outcome  compared  with  the  latter.  Accordingly, the beneficial effects of propofol ob-served  could  not  completely  account  for  the  clot  volume  reduction  or  hemodynamic  stabilization.  Other  propofol-induced  pharmacological  actions  such  as  direct  neuroprotection  should  further  be  evaluated in detail.

Here, patients who received propofol sedation at a dose > 3.33 mg/kg/hr in the first 12 hours after surgery  had  a  significantly  larger  recurrent  blood  clot  in  the  brain  after  craniotomy  for  hematoma  evacuation (p < 0.05) than the other two groups of sedated  patients.  Interesting,  Aoki  et  al  reported  that  propofol  inhibited  platelet  aggregation  both  in  vivo  and  in  vitro.²²  Beule  et  al  also  reported  that  platelet  function  and  a  significant  propofol  anesthesia  showed  interaction  in  patients  under-going  endoscopic  sinus  surgery.²³  It  was  possible  that the use of high dose propofol (> 3.33 mg/kg/hr)  might  induce  an  adverse  effect  of  propofol  on  platelet  function,  subsequently  resulting  in  a  larger  recurrent  blood  clot  and  the  loss  of  the  propofol-induced  beneficial  effect  as  observed  in  patients  with  the  optimal  dosage.  However,  fur-ther  studies  are  needed  to  evaluate  this  possible  detrimental  effect  on  platelet  function  and  its  relation to dosage.

In  conclusion,  we  have  demonstrated  that  the  use  of  propofol  sedation  during  the  early  postop-erative period could benefit neurologic and clinical outcomes by improving GCS scores and decreasing the  30-day  mortality  rate  in  hemorrhagic  stroke  patients  following  a  decompressive  craniotomy.  However, early postoperative use of propofol seda-tion  at  large  doses  might  warrant  special  atten-tion, because it tended to induce greater residual/recurrent   hematoma   formation   after   surgery.   Further studies are needed to evaluate its possible detrimental effect on platelet function, especially when  it  is  used  at  a  large  dose  during  the  early  postoperative period.