Background: Despite meticulous hemostasis, persistent postoperative drain output following posterior cervical spine procedures often necessitates a prolonged length of hospital stay. We sought to determine if thrombin-soaked absorbable gelatin compressed sponge can decrease postoperative drain output and the length of hospital stay after multilevel posterior cervical spine surgery.
Methods: We performed a retrospective analysis of forty-three pairs of patients who had undergone either posterior cervical decompression and/or fusion of three or more levels by the same surgeon. The patients were matched according to intraoperative blood loss, age, sex, and number of involved levels. Control patients were managed between 2004 and 2007, whereas study patients were managed between 2008 and 2011. The only variable between the study and control groups was that, in the study group, absorbable gelatin compressed sponge was soaked in thrombin and applied over the exposed spine before wound closure. A subfascial drain was used in all patients. Total drain output, time for the drainage to decrease to <30 mL per eight-hour shift (at which point the drain was discontinued), the length of stay, the number of readmissions, and postoperative complications were analyzed.
Results: Total drain output averaged 93 mL in the study group and 204 mL in the control group (p < 0.0001). The average time for the drainage to decrease to <30 mL per eight-hour shift was 2.5 shifts in the study group and 4.4 shifts in the control group (p < 0.0001). Length of stay averaged 1.3 days (cumulative total, fifty-seven days) in the study group and 2.2 days (cumulative total, ninety-five days) in the control group (p < 0.0001). Persistent drain output was the primary reason preventing discharge on the first postoperative day. There were no infections, epidural hematomas, or readmissions within thirty days of discharge in either group. No patient developed adverse reactions attributable to the thrombin-soaked absorbable gelatin compressed sponge.
Conclusions: Application of thrombin-soaked absorbable gelatin compressed sponge at the end of multilevel posterior cervical spinal surgery significantly decreased postoperative drain output and consequent hospital stay.
Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.
1Department of Orthopaedics, Mount Sinai School of Medicine, 5 East 28th Street, Box 1188, New York, NY 10029
2Department of Orthopaedic Surgery, Washington University, 660 South Euclid Avenue, Campus Box 8233, St. Louis, MO 63110. E-mail address for K.D. Riew: email@example.com
3Department of Orthopaedic Surgery, Medical College of Hallym University, 896, Pyeongchon-dong, Dongan-gu Anyang-si, Gyeonggi-do 431-070, South Korea
4Department of Orthopaedic Surgery, Asan Medical Center, University of Ulsan, 388-1 Pungnap 2-dong, Songpa-gu, Seoul 138-736, South Korea
Persistent postoperative bleeding following posterior cervical spinal surgery can lead to prolonged hospitalization and may result in the formation of epidural hematoma that can cause spinal cord compression and consequent neurologic deficits in rare cases1-5. The use of a subfascial drain after multilevel posterior cervical procedures may prevent fluid collection and formation of a large hematoma. Continual drain output, however, increases the length of hospital stay and often is the primary reason for delaying patient discharge beyond the first postoperative day.
Numerous hemostatic agents with different mechanisms of action are currently available for use during surgery to control bleeding. Absorbable gelatin compressed sponge (Gelfoam; Pharmacia and Upjohn, Kalamazoo, Michigan) is one such product intended for the application to bleeding surfaces as an effective hemostatic agent6-12. It is a pliable, hydrophilic sponge prepared from purified porcine skin gelatin granules, and its mode of action is mostly physical rather than an alteration of the blood-clotting mechanism. In contrast, thrombin, also known as coagulation factor II, is a serine protease coagulation protein that is a well-known natural hemostatic. Its main mechanism of action is conversion of soluble fibrinogen into insoluble strands of active fibrin during clot formation.
In the present study, we sought to take advantage of both the mechanical and chemical hemostatic properties afforded by the simultaneous use of thrombin and absorbable gelatin compressed sponge, respectively, and hypothesized that the application of thrombin-soaked gelatin sponge would reduce postoperative bleeding and subsequent drain output following multilevel posterior cervical spinal surgery. Furthermore, we hypothesized decreased drain output would lead to shortened hospital length of stay. To that end, we conducted a retrospective matched-pair analysis.
Materials and Methods
After approval was obtained from the institutional review board, a retrospective review of consecutive patients who had undergone multilevel posterior cervical procedures (procedures involving three levels or more) between January 2004 and June 2011 was performed. Patients were divided into two groups: those who received thrombin-soaked gelatin sponge (managed between January 2008 and June 2011) and controls (managed between January 2004 and December 2007). To minimize other variables such as changes in surgical techniques and the use of other hemostatic agents, we decided that all of the controls would be patients who had been managed during the four years immediately before the senior surgeon (K.D.R.) started using thrombin-soaked gelatin sponge for all multilevel posterior cervical procedures.
The inclusion criteria were (1) posterior cervical decompression and/or fusion of three or more levels (i.e., motion segments) between C1 and the cephalad aspect of the thoracic spine (down to T4), (2) cervical spondylosis with or without disc herniation causing radiculopathy and/or myelopathy, (3) pseudarthrosis resulting from previous anterior or posterior cervical fusion, (4) adjacent-segment breakdown resulting from previous anterior or posterior cervical fusion, (5) congenital spinal stenosis, and/or (6) ossification of the posterior longitudinal ligament. The exclusion criteria included (1) trauma, (2) infection, (3) tumors, (4) one or two-level procedures, (5) osteotomies, (6) procedures extending in a cephalad direction to the occiput or in a caudad direction to the midthoracic spine (T5 and down), (7) bleeding diathesis or recent history of anticoagulation, and (8) pregnancy.
One hundred and one patients in the control group (including fifty-six patients managed with arthrodesis and forty-five managed with decompression only) and 137 patients in the study group (including eighty-four managed with arthrodesis and fifty-three managed with decompression only) fulfilled the inclusion and exclusion criteria. We performed 1:1 matching on the basis of intraoperative blood loss (±50 mL), age (within five years), sex, and the number of decompression and/or fusion levels (within one level). Control patients who had had arthrodesis were matched to study patients who also had had arthrodesis. Likewise, control patients who had had decompression were only paired with study patients who had had decompression. We matched forty-three pairs of patients (including sixteen pairs of patients who had had arthrodesis and twenty-seven pairs of patients who had had decompression only). Patient selection and analysis were conducted by independent spine surgeons who were not involved in patient care.
Application of Thrombin-Soaked Absorbable Gelatin Compressed Sponge
All procedures were performed by a single surgeon at one hospital. In both the study and control groups, meticulous hemostasis was achieved at the end of the procedure with use of bipolar electrocautery, hemostatic matrix (Surgiflow; Johnson & Johnson, Somerville, New Jersey), and demineralized bone matrix soaked in thrombin, with particular attention to the control of bleeding from the exposed bone and epidural vessels. For the fusion cases, local autograft was used along with allograft and bone graft extenders (e.g., demineralized bone matrix, rhBMP-2 [recombinant human bone morphogenetic protein-2]) on an as-needed basis. In the experimental group, after adequate hemostasis had been achieved, a single piece of absorbable gelatin compressed sponge (size, 100 cm2) was soaked in thrombin (2500 IU) (Gelfoam Plus; Baxter Healthcare, Deerfield, Illinois) (Fig. 1). In cases in which the wound was small, we cut the piece such that the entire wound would be covered. In cases in which the wound was bigger than the single piece, we cut the piece longitudinally in half and placed the two pieces in the middle of the wound, on top of the bone graft material. We did not place the sponge over the exposed dura, as it can expand in the wound and compress the spinal cord underneath. In both the study and control groups, a deep drain (Hemovac; Zimmer, Warsaw, Indiana) was placed below the fascia and the wound was closed in multiple layers. A second superficial Hemovac drain was also placed above the fascia in a small number of patients. The drained fluid was initially sanguinous and gradually became serous after two to three eight-hour shifts.
Measurement of Postoperative Drain Output and Criteria for Discharge
All data, including the drain amount, were obtained from the electronic inpatient medical records. The drain output was measured and recorded three times a day in eight-hour shifts (at 7 A.M., 3 P.M., and 11 P.M.). When there were two drains, one deep and one superficial to the fascia, the output amounts were combined together. The drain was routinely removed when the drain output per eight-hour shift was <30 mL. Unless there were contraindications to discharge (e.g., medical complications or comorbidities), the patient was discharged once the drain was removed.
Comparison Between Study and Control Groups
Four parameters were compared between the two groups: (1) total drain output before the removal of the drain, (2) time for the drainage to decrease to <30 mL per eight-hour shift, (3) length of hospital stay (measured in days), and (4) number of patients having gelatin sponge-related or any other complications requiring readmission within one month after the index procedure.
Distributions of variables were given as the mean and the standard deviation. The Student t test was used to assess the difference of continuous measures between the groups. The Fisher exact test was used for dichotomous data analysis. The level of significance was set at p < 0.05.
Source of Funding
There was no external funding source.
One hundred and thirty-seven patients in the experimental group met the inclusion and exclusion criteria. Of these, eighty-four patients underwent arthrodesis with or without decompression. Fifty-three patients underwent decompression only. There were 101 patients in the control group, of whom fifty-six underwent arthrodesis and the remaining forty-five had decompression alone. After matching of the patients according to intraoperative blood loss, age, sex, number of levels, and procedure (arthrodesis versus decompression alone), forty-three pairs of patients were included in this study. There were sixteen pairs of patients in the arthrodesis category and twenty-seven pairs in the decompression-only category (see Appendix).
Total Drain Output
The total drain output before drain removal averaged 204.4 mL (range, 20 to 500 mL) in the control group, compared with 92.9 mL (range, 0 to 380 mL) in the study group (p < 0.0001). Similar trends were observed when patients were subanalyzed according to whether they had been managed with decompression only (mean, 181.7 mL in the control group versus 54.8 mL in the study group; p < 0.0001) or arthrodesis (mean, 242.8 mL in the control group versus 150.1 mL in the study group; p = 0.0345) (Fig. 2).
Time for Drainage to Decrease to <30 mL per Eight-Hour Shift
The time for the drainage to decrease to <30 mL per shift averaged 35.2 hours (range, eight to sixty-four hours) in the control group, compared with 19.7 hours (range, eight to forty hours) in the experimental group (p < 0.0001). Similar trends were seen in the subanalysis of the patients who had been managed with decompression only (mean, 32.9 hours in the control group versus 14.4 hours in the study group; p < 0.0001) or arthrodesis (mean, 39.0 hours in the control group versus 28.0 hours in the study group; p < 0.0316) (Fig. 3).
Length of Stay
The average length of hospital stay was 2.2 days (range, one to five days, for a cumulative total of ninety-five days) in the control group, compared with 1.3 days (range, one to three days, for a cumulative total of fifty-seven days) in the study group (p < 0.0001) (Fig. 4). Similar trends were observed in the subanalysis of patients managed with decompression only (mean, 2.1 days in the control group versus 1.3 days in the study group; p < 0.0001) or arthrodesis (mean, 2.3 days in the control group versus 1.4 days in the study group; p = 0.0007). For one patient in the control group, hospital discharge was delayed by a medical complication (pulmonary embolus). In this case, the drain was discontinued two days after surgery, but the patient was discharged on the fifth postoperative day once warfarin treatment had reached therapeutic levels.
Figure 5 illustrates that the majority of patients in the study group (thirty-one of forty-three) left the hospital on the first postoperative day, with ten leaving on the second postoperative day and two leaving on the third postoperative day. In contrast, discharge was more dispersed for patients in the control group, with only six patients leaving the hospital on the first postoperative day, twenty-five leaving on the second postoperative day, ten leaving on the third postoperative day, one leaving on the fourth postoperative day, and one remaining until the fifth postoperative day because of a medical complication.
No patient in either group was readmitted because of postoperative complications within the first thirty days after surgery. There were no infections or clinically identifiable epidural hematomas. There were no adverse reactions attributable to the thrombin-soaked absorbable gelatin compressed sponge.
Prolonged postoperative bleeding following multilevel posterior cervical spinal procedures may lengthen hospitalization and, in rare but the most severe cases, may cause symptomatic epidural hematoma formation, leading to spinal cord compression and subsequent neurologic deficits1-5. Persistent drainage often is the main reason for delayed patient discharge on the first postoperative day. In the current study, we performed a retrospective matched-pair analysis to demonstrate an easy, safe, and cost-effective method to reduce postoperative drainage by using a combination of existing and easily available hemostatic agents.
Yeom et al. recently reported on the application of fibrin sealant (TISSEEL; Baxter) at the end of multilevel anterior cervical arthrodesis as an effective method of decreasing postoperative drain output and length of hospital stay13. Unlike anterior procedures, posterior cervical procedures require the dissection of multiple layers of paraspinal muscles that may become the major source of persistent bleeding following surgery. Therefore, while fibrin sealants are highly successful for sealing the site of drainage (for example, leakage of cerebrospinal fluid from a dural tear), they are often ineffective when there is active bleeding. In contrast, absorbable gelatin compressed sponge (Gelfoam) is a topical hemostatic device intended to be used on bleeding surfaces. Furthermore, the hemostatic property of gelatin sponge is enhanced when used in conjunction with thrombin.
When thrombin-soaked gelatin sponge was used, the total drain output significantly decreased from 204.4 to 92.9 mL (p < 0.0001). Moreover, the time for the output to decrease to <30 mL per shift decreased from 35.2 to 19.7 hours (p < 0.0001). Consequently, as seen in Figure 5, the majority of patients left the hospital on the first postoperative day, even after posterior procedures involving three to seven levels.
The combined gelatin sponge and thrombin (Gelfoam Plus) costs approximately $129 for a multilevel procedure of three or more levels, whereas one day of hospitalization costs approximately $100013. The cumulative length of hospital stay was ninety-five days for the patients in the control group, compared with fifty-seven days for the patients in the study group. The use of thrombin-soaked gelatin sponge decreased the number of hospital days by thirty-eight, yielding a net savings of >$33,000 for just forty-three patients. This simple and safe method of decreasing postoperative bleeding and the consequent length of hospital stay may be one way to be cost-effective following multilevel posterior cervical procedures.
Although the current study was not a prospective randomized trial and was associated with the potential weaknesses inherent in retrospective analysis, it included patients who were matched on the basis of intraoperative blood loss, age, sex, and number of involved levels. Demographic and surgical parameters were comparable between the two groups. A limitation of the study is that the reviewers were not blind to the use of thrombin-soaked absorbable gelatin compressed sponge. However, all of the collected data, including the time of discharge, the amount of drain output, and the length of hospital stay, are unalterable and are not subject to any interpretation error. Furthermore, the data were collected and analyzed by surgeons who were uninvolved in the care of the patients.
Another potential weakness is the use of <30 mL per eight-hour shift as the cutoff for discontinuing the drain. There is no standard protocol for drain usage following extensive posterior cervical spinal surgery as its use is often a matter of personal choice at the discretion of the surgeon. In a prospective, randomized study, Mirzai et al. detected epidural hematomas in 36% of patients who were managed with a drain following lumbar disc surgery and in 89% of patients who were managed without a drain (p = 0.000)14. Although the incidence of symptomatic epidural hematoma that necessitates evacuation is quite low (0.1% to 3%)15 and we are not aware of any published level-I or II evidence that clearly delineates the benefit of using a drain, we believe that using a subfascial drain is a safe and effective means of decreasing postoperative epidural hematoma, especially after multilevel surgery, which is a substantial risk factor for the development of symptomatic epidural hematoma3. As for the timing of drain removal, the authors of one study noted that surgeons removed the drain if the output was <50 mL per twelve hours15. The criterion to discontinue the drain when there was <30 mL per shift was derived from the experience of the senior author (K.D.R.), who noted a higher rate of wound complications when the drain was pulled earlier.
The length of hospital stay following orthopaedic procedures generally has been decreasing over the past decade16-18. At our institution, there has been a significant decrease in the length of hospital stay following total hip arthroplasty (from 3.6 days [in 2004 to 2007] to 2.5 days [in 2008 to 2011]; p = 0.0140) and total knee arthroplasty (from 3.4 to 2.6 days; p = 0.0045), but not after shoulder arthroplasty (from 2.4 to 2.3 days; p = 0.4728). However, the reported length of hospital stay during the same period following multilevel posterior cervical procedures remains approximately four days and does not seem to have decreased significantly19,20. The retrospective nature of our analysis leaves room for other unidentified changes that may have occurred over the course of the data-collection period to confound the results of our study. In patients who were medically and neurologically stable, the only factor keeping the patient in the hospital was the maintenance of the drain. A significant decrease in drain output and consequent length of hospital stay following the use of thrombin-soaked gelatin sponge in our patients appears to support the application of this hemostatic agent following posterior cervical arthrodesis and decompression procedures.
Tables showing demographic and clinical data on the study and control groups are available with the online version of this article as a data supplement at jbjs.org.
Investigation performed at the Cervical Spine Service, Washington University School of Medicine, St. Louis, Missouri
1. Sokolowski MJ Dolan M Aminian A Haak MH Schafer MF. Delayed epidural hematoma after spinal surgery: a report of 4 cases. J Spinal Disord Tech. 2006 Dec;19(8):603–6.
2. Neo M Sakamoto T Fujibayashi S Nakamura T. Delayed postoperative spinal epidural hematoma causing tetraplegia. Case report. J Neurosurg Spine. 2006 Sep;5(3):251–3.
3. Kou J Fischgrund J Biddinger A Herkowitz H. Risk factors for spinal epidural hematoma after spinal surgery. Spine (Phila Pa 1976). 2002 Aug 1;27(15):1670–3.
4. Glotzbecker MP Bono CM Wood KB Harris MB. Postoperative spinal epidural hematoma: a systematic review. Spine (Phila Pa 1976). 2010 May 1;35(10):E413–20.
5. Yi S Yoon do H Kim KN Kim SH Shin HC. Postoperative spinal epidural hematoma: risk factor and clinical outcome. Yonsei Med J. 2006 Jun 30;47(3):326–32.
6. Guralnick WC Berg L. Gelfoam in oral surgery; a report of 250 cases. Oral Surg Oral Med Oral Pathol. 1948 Jul;1(7):632–9.
7. Barnes AC. The use of gelatin foam sponges in obstetrics and gynecology. Am J Obstet Gynecol. 1947 Jul;54(1):105–7.
8. Jenkins HP Clarke JS. Gelatin sponge, a new hemostatic substance; studies on absorbability. Arch Surg. 1945 Nov-Dec;51:253–61.
9. Jenkins HP Janda R Clarke J. Clinical and experimental observations on the use of gelatin sponge or foam. Surgery. 1946 Jul;20(1):124–32.
10. Jenkins HP Janda R. Studies on the use of gelatin sponge or foam as an hemostatic agent in experimental liver resections and injuries to large veins. Ann Surg. 1946 Nov;124:952–61.
11. Jenkins HP Senz EH . Present status of gelatin sponge for the control of hemorrhage; with experimental data on its use for wounds of the great vessels and the heart. J Am Med Assoc. 1946 Nov 16;132(11):614–9.
12. Jenkins HP Jampolis R . Control of hemorrhage by gelatin sponge in experimental wounds of the great vessels and the heart. Proc Inst Med Chic. 1947 May 15;16(14):403.
13. Yeom JS Buchowski JM Shen HX Liu G Bunmaprasert T Riew KD. Effect of fibrin sealant on drain output and duration of hospitalization after multilevel anterior cervical fusion: a retrospective matched pair analysis. Spine (Phila Pa 1976). 2008 Jul 15;33(16):E543–7.
14. Mirzai H Eminoglu M Orguc S. Are drains useful for lumbar disc surgery? A prospective, randomized clinical study. J Spinal Disord Tech. 2006 May;19(3):171–7.
15. Aono H Ohwada T Hosono N Tobimatsu H Ariga K Fuji T Iwasaki M. Incidence of postoperative symptomatic epidural hematoma in spinal decompression surgery. J Neurosurg Spine. 2011 Aug;15(2):202–5. Epub 2011 May 6.
16. Vorhies JS Wang Y Herndon JH Maloney WJ Huddleston JI. Decreased length of stay after TKA is not associated with increased readmission rates in a national Medicare sample. Clin Orthop Relat Res. 2012 Jan;470(1):166–71.
17. Vorhies JS Wang Y Herndon J Maloney WJ Huddleston JI. Readmission and length of stay after total hip arthroplasty in a national Medicare sample. J Arthroplasty. 2011 Sep;26(6 Suppl):119–23. Epub 2011 Jul 1.
18. Husted H Jensen CM Solgaard S Kehlet H. Reduced length of stay following hip and knee arthroplasty in Denmark 2000-2009: from research to implementation. Arch Orthop Trauma Surg. 2012 Jan;132(1):101–4. Epub 2011 Sep 24.
19. Highsmith JM Dhall SS Haid RW Jr Rodts GE Jr Mummaneni PV. Treatment of cervical stenotic myelopathy: a cost and outcome comparison of laminoplasty versus laminectomy and lateral mass fusion. J Neurosurg Spine. 2011 May;14(5):619–25.
20. Heller JG Edwards CC 2nd Murakami H Rodts GE. Laminoplasty versus laminectomy and fusion for multilevel cervical myelopathy: an independent matched cohort analysis. Spine (Phila Pa 1976). 2001 Jun 15;26(12):1330–6.
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.