Incidence of deep venous thrombosis in patients undergoing elective neurosurgery – A prospective cohort based study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/neuroindia.NI_1237_15
Source of Support: None, Conflict of Interest: None
Keywords: Doppler ultrasound, deep vein thrombosis, elective neurosurgery, risk factors
The term venous thromboembolism (VTE) encompasses both deep vein thrombosis (DVT) and its serious complication, pulmonary embolism (PE). Neurosurgical patients are considered to be under moderate risk for DVT. Various factors place the neurosurgery population at an increased risk for DVT; these factors include malignancy, duration of surgery, decreased mobilization postoperatively, postoperative motor deficits, and increasing age. The incidence of DVT in neurosurgical patients (cranial and spinal combined) is reported to be 29–43% without any pharmacological prophylaxis, with only 25% of patients becoming clinically symptomatic., The risk of pulmonary embolism (PE) in neurosurgical patients is 5% with the mortality ranging from 9 to 50%., It is also known that DVT may be asymptomatic and its initial presentation may be that of fatal PE. Sixty to eighty percent of fatal PE are unsuspected and undiagnosed. Screening is therefore important. There are no clear guidelines for the prophylaxis of DVT in neurosurgery. The reason for this dilemma may be the high risk of intracranial bleeding following pharmacological prophylaxis for DVTi n the neurosurgical population. According to the recent American College of Chest Physicians (ACCP) 2012 guidelines, mechanical prophylaxis is preferred over pharmacological prophylaxis in routine craniotomy patients.
Even though Indians may be at an equal risk, the exact incidence of DVT is not known for the Indian population. There are only a handful of studies, mainly orthopedic, in which the rate of DVT varies from 3.7 to 17%., In the present scenario, a standard guideline for providing thromboprophylaxis to patients in the Indian subcontinent is needed, which should be practical and acceptable for all. For this, we need data profiling only Indians because there are bioracial differences between the Caucasian and the other ethnic groups. There are no published Indian studies documenting the incidence of DVT in neurosurgical patients. This study was thus conducted to determine the incidence of DVT in Indian patients undergoing standard neurosurgical procedures. This is a first single institutional cohort based study to determine the incidence of postoperative DVT in the Indian population. We hope to assess the demographic and clinicopathological factors contributing to DVT and document its incidence and prevalence. We also hope that this study can provide guidelines for appropriate and timely prophylaxis of DVT.
This prospective study included patients who underwent elective neurosurgery during the period of one year at NIMHANS from December 2013 to November 2014. A prior informed consent was obtained from all the patients and the study was approved by the Ethical Committee of the Institute. All patients underwent baseline evaluation of coagulation parameters prior to their surgery and were screened with a preoperative ultrasound color Doppler imaging to rule out DVT in the preoperative period. All the patients underwent assessment of both the lower limbs on the 3rd and 10th postoperative day. Serial evaluation was done subsequently, weekly, till the discharge/death of the patient to document any progression of the thrombus. The study group included all patients who developed DVT in the postoperative period and the control group included all patients without DVT.
The color Doppler ultrasound utilized the 5–10 MHz linear transducer in the real time B mode. The Doppler assessment included examination of the bilateral common femoral, superficial femoral, popliteal, anterior tibial and posterior tibial veins in both longitudinal and transverse planes in both the limbs. They were assessed for flow, visualized thrombus, compressibility, and augmentation. The results were assessed by an experienced radiologist. A diagnosis of DVT was made when there was visualization of the thrombosis, absence of flow, lack of compressibility, or lack of augmentation. The thrombus was classified as distal if it involved the calf veins only, and as proximal if it involved the popliteal or a more proximal vein. Postoperative coagulation profile was obtained on the 3th or 4th day. The patients who developed postoperative DVT, diagnosed by Doppler examination, were subjected to thromboprophylaxis as per the institute protocol.
Inclusion and exclusion criteria
The inclusion crtieria included all patients >18 years of age who underwent elective neurosurgery at NIMHANS. The exclusion criteria were: 1) Patients >18 years of age, 2) patients with preoperative DVT, and those with a known history of coagulation disorders, 3) patients undergoing surgery (including the anaesthesia time) lasting for less than 4 hours, 4) patients with subaxial spine injuries, 5) patients with a likely hospital stay of less than 48 hours, 6) all patients with traumatic injury, and 7) patients with refusal of consent.
The Statistical Package for the Social Science version 15.0 (SPSS15.0 program, SPSS, Inc. Chicago, IL) was used for data analysis. The percentages were compared using the chi-square test. P<0.05 was considered significant. In addition, Pearson's correlation analysis was used to determine whether or not significant correlations existed between the chosen variables.
A total of 273 patients who met the inclusion criteria were included in the study, out of which 33 patients (12.08%) developed DVT. The patients were then analyzed using various subgroup analyses, which is detailed below.
There were a total of 171 male (62.6%) and 102 female (37%) patients. Among patients who developed DVT, the female subjects constituted 57.6% and the males ones, 42.4%. In the DVT group, the mean age was 43 years, ranging from 19 to 70 years.
Age: All the patients in the study were divided into 3 age groups (<40 years, 40–60 years and >60 years). The incidence of DVT was analyzed across these 3 age groups. There was a significant difference between the incidence of DVT in the 40–60 age group compared to those who did not develop DVT (P = 0.012). The incidence of DVT did not increase significantly beyond the age of 60 years (P = 0.012).
Gender: Male subjects constituted 62% and 65% of patients in the DVT and non-DVT groups, respectively. 37.4% female subjects developed DVT and 34.6% did not develop DVT. There was a significant increase in the incidence of DVT among the female as compared to the male (P = 0.025) patients.
Comorbidities: Hypertension was significantly associated with increased incidence of DVT (P = 0.06). Frequent alcohol intake, smoking, and obesity were found to be the risk factors for DVT (P = 0.00) [Figure 1].
Preoperative Glasgow coma scale: The neurological status of the patients was recorded according to the Glasgow coma scale (GCS). In the DVT group, the mean preoperative GCS score was 13 (minimum, 0 and maximum, 15); and, in the non-DVT group, the mean preoperative GCS was 14 (minimum, 0 and maximum, 15).
Preoperative motor deficits: The occurrence of preoperative motor deficits did not significantly influence the occurrence of DVT in the postoperative period (P = 0.07).
Diagnosis: Tumors were the most common overall pathology. Glioblastoma were the most common malignant tumor (54%) followed by anaplastic oligodendrogliomas (45%). Meningiomas were the most common benign tumours (66%) followed by pituitary adenomas (18%). A significant association was observed between malignant tumors as well as meningiomas, and the development of DVT in the postoperative period (P = 0.001). No significant association was observed between aneurysmal subarachnoid hemorrhage and the development of DVT (P = 0.16) [Figure 2].
Position during surgery: All patients undergoing surgery were positioned in the supine, lateral or prone position, as per the requirement of the particular surgery. In the DVT group, almost 91% patients were in the supine position. 6.1% of the patients undergoing surgery in the lateral position developed DVT.A significantly high proportion of DVT was observed in patients undergoing surgery in the supine and lateral position when compared to the prone position (P = 0.001).
Duration of surgery: Most patients underwent surgery for atleast 5 hours. 91% patients in the DVT group had undergone surgery for more than 5 hours. In the non-DVT group, 75% patients had surgery lasting for >5 hours. Patients undergoing surgery for more than 5 hours were observed to have a significant risk of developing DVT as compared to those whose duration of surgery was less than 5 hours. (P = 0.001)
Post-operative motor deficits: Postoperative motor deficits were quantified using the Medical Research Council grading. In the DVT group, 88% of patients had motor deficits; whereas, in the non-DVT group, 16% had motor deficits. The presence of postoperative motor deficits was significantly associated with the occurrence of DVT (P = 0.001) [Figure 3].
Severity of motor deficit: In the DVT group, 54% patients had a deficit with power less than 3, approximately 33% had a deficit with power more than 3, and 12% had full grade power of 5. In the non-DVT group, 5% had power less than 3, 10% had power greater than 3, whereas 84% had no deficits. Among patients with postoperative DVT, 45.5% had left hemiparesis, 30% had right hemiparesis, whereas 12% had bilateral limb weakness. The severity of motor deficits after surgery significantly influenced the occurrence of DVT in the postoperative period (P = 0.01).
Among the 33 patients with DVT, 69.7% were asymptomatic and were detected on surveillance ultrasound Doppler examination, whereas 30% had symptoms related to DVT including swelling, redness, and tenderness of the affected limb. Thus, a significant proportion of asymptomatic DVT was diagnosed on surveillance alone (P = 0.0001).
Time-to-ambulation period: In the DVT group, 94% patients had a time-to-ambulation (TTA) period of more than 2 days, whereas in the non-DVT group, only 22% patients had a TTA period of more than 2 days. The TTA period of more than 2 days was significantly associated with the development of DVT in the postoperative period (P = 0.001) as compared to patients who were ambulated within 2 days of the surgery.
Extent and location of deep vein thrombosis: 51% of the DVT occurred in the left lower limb, 36% in the right lower limb, and 12% in both the lower limbs. The DVT extended along both the proximal and distal deep venous system in most of the patients (84%). Isolated distal DVT was observed in 12.2% of patients. 3% of patients had DVT involving the proximal deep veins only [Table 1].
Intensive care unit and ventilatory support: Among the patients with DVT, 54.5% needed intensive care unit (ICU) admission with ventilatory support. In the non-DVT group, almost 89% of the patients did not need an ICU or ventilatory support. However, no significant difference was observed between the necessity of ICU care and ventilatory support in the postoperative period and the development of DVT (P = 1.01).
Hospital stay: The total hospital stay was calculated from the time of surgery until discharge from the hospital or death in the hospital. In the DVT group, the mean hospital stay was 20.51 days (range: 8–62 days), and in the non-DVT group, it was 16.19 days (range: 17–117 days). Thus, the development of DVT significantly increased the mean hospital stay of the patients (P = 0.001).
Glasgow coma scale at discharge: The sensorium of all included patients was documented according to the GCS score. 85% of the patients with DVT had a GCS score of 15 at discharge. The presence of postoperative DVT did not significantly influence the sensorium of the patient at discharge (P = 0.112).
Management of deep vein thrombosis: Anticoagulation was initiated in all patients diagnosed with DVT. Initially, injectable heparin was started (unfractionated in 51.5% and low molecular weight in 48.5%). Oral anticoagulation with warfarin was overlapped with injectable heparin after 48–72 hours in all patients. The therapeutic range of anticoagulation was maintained with serial monitoring of international normalised ratio (INR).
Follow up: Two-third of the patients were followed-up after discharge. Follow-up Doppler imaging was performed on all these patients. The mean follow-up duration was 6.5 months (minimum, 3 and maximum, 10 months).27% of patients demonstrated persistent DVT at follow-up. All these patients had symptomatic DVT at the initial diagnosis. Resolution of DVT was observed in 73% of the patients at follow-up.
In our study, the proportion of patients who developed DVT in the postoperative period was 12.08% (33/273 patients). Leizorovicz et al., have reported a frequency of acute DVT of approximately 16 to 44% and 15.8%, in hip and knee replacement and hip fracture surgeries, respectively, in the Asian population. In the western literature, the reported incidence of postoperative DVT in neurosurgical patients varied from 3 to 25%.,, Our study reports an incidence of DVT comparable to the other Asian countries, although not as high as that in the western population [Table 2].,,
The proportion of DVT was significantly high in the 40–60 year age group as compared to that occurring in patients below 40 years and those beyond 60 years of age. Old age beyond 65 years is a known risk factor.,, The variable results in our study may be due to a small sample size. In our study, the proportion of DVT was higher in female than that in male subjects. Most western studies have reported a higher incidences in males., However, in the Indian population, a significant difference between males and females has not been reported., Hypertension increased the proportion of DVT. In the literature, a two-fold increased risk of developing DVT in patients with hypertension has been described.,,, The occurrence of preoperative motor deficits did not significantly influence the occurrence of DVT in the postoperative period, which is in contrary to the observations published by Quevedo et al., who observed that a preoperative paretic extremity increased the risk of DVT. Gliomas and meningiomas are known to aggravate prothrombotic events. A significant association was observed between malignant gliomas as well as meningiomas, and the development of DVT in the postoperative period, and these findings are similar to that reported by Semrad et al., and Brandes et al.,
In our study, a high proportion of DVT was observed in patients undergoing surgery in the supine and lateral position and in whom surgery lasted for more than 5 hours. These are similar to a few studies in literature., Raslan et al., and Semrad et al., identified surgery duration greater than 4 hours to be a major risk factor., Intraoperative blood transfusion of more than 5 units increased the incidence of postoperative DVT. Subclinical coagulopathy can occur in massive blood transfusions and may increase the risk of venous thrombosis.
Limb weakness is a well-known risk factor. In our study, the presence of postoperative motor deficit and its severity was significantly associated with the occurrence of DVT. Our findings are comparable with that in literature that have reported postoperative hemiparesis and paraparesis as a significant risk factor., Delayed ambulation aggravates the venous stasis and thereby increases the risk for DVT. In this study, a TTA of more than 2 days resulted in an increased risk of DVT in the postoperative period, which is similar to the results reported by Ray et al., and Bagaria et al.
In the study group, 69.7% of asymptomatic patients with DVT were detected on surveillance ultrasound Doppler examination, a figure that is higher than that reported by Dermody et al. The proportion of asymptomatic DVT detected was significantly high among those screened for it. Majority of the patients with asymptomatic DVT had both proximal and distal DVT. The proportion of isolated proximal DVT in our study was lower as compared to the literature. The development of DVT significantly increased the mean hospital stay, which is in unison with previous studies. Prolonged stay in the ICU in postoperative period is often associated with an increase in procoagulant factors with a simultaneous decrease in the anticoagulant factors that increase the risk of DVT. Though our study did not report any statistically significant association between the length of ICU stay and the development of DVT, we do recommend routine screening in patients with prolonged ICU hospitalization. Based on our experience, we suggest an algorithm which can prove to be useful in patients undergoing elective neurosurgery [Figure 4].
This is a single institutional study, one of the first of its kind, documenting the incidence of DVT in Indian patients. Our results were in consonance with those in the Western literature; however there were a few important differences. The incidence of DVT was lower than that seen in the Western literature, and also a high proportion of DVT was asymptomatic. We, thus, emphasize the utility of Doppler ultrasound as a surveillance screening tool and recommend the usage of prophylaxis (mechanical/pharmacological) in high risk patients, as detailed. We believe this study provides useful insights into the understanding of the independent perioperative factors that predispose to VTEs and will help to guide treatment strategies aimed at minimizing VTE in Indian neurosurgical patients.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2]