Abstract
Introduction
Although previous studies reported the clinical significance of drains in lumbar surgery, their role in and effects on the clinical outcomes of cervical spine surgery remain unclear. The present study compared the clinical outcomes of cervical laminoplasty (CLP) using a closed suction drain (CSD) and closed nonsuction drain (CNSD).
Methods
Prospectively recorded surgical data on consecutive patients who underwent CLP at a single institution between 2014 and 2020 and were followed up for at least 1 year were examined. CSD was used prior to January 2018, and CNSD has since been employed. One hundred patients who underwent surgery before and after the change in drain type (the CSD and CNSD groups, respectively) were selected for analysis. Primary outcome measures were the drainage amount, blood count, and fluid collection at the surgical site defined by magnetic resonance images. The Japanese Orthopaedic Association (JOA) score for the cervical spine was also evaluated as a functional outcome.
Results
No significant differences were observed in demographic, baseline clinical, or surgical data between the CSD and CNSD groups. The drainage amount was significantly greater in the CSD group than in the CNSD group (224 vs. 143 mL, P<0.001). Hemoglobin and hematocrit levels were significantly decreased in the CSD group than in the CNSD group. Medium or large fluid collection was significantly more common in the CNSD group than in the CSD group. No significant differences were observed in the number of surgical site infections, the formation of symptomatic hematoma, or JOA scores between the two groups.
Conclusions
The use of CNSD in CLP decreased the drainage amount and maintained the hemoglobin level compared with that of CSD. Although no patients developed symptomatic hematoma, the amount of epidural fluid collected was larger in the CNSD group than in the CSD group.
Keywords: drain, closed suction drain, closed non-suction drain, cervical laminoplasty, hematoma, surgical site infection, complication
Introduction
Closed suction drains (CSD) have been used in spinal surgery to prevent the formation of postoperative hematomas and infections1). However, this is controversial because the number of complications associated with the use of drains is high2). Nevertheless, many surgeons continue to customarily use drains3). Randomized control studies (RCTs) were recently performed, and evidence was obtained on the use of drains at the lumbar level4,5); however, limited information is currently available on the effects of using or not using drains in posterior surgery on the cervical spine, which may be attributed to suction drains being employed to reduce the potential risk of spinal cord damage caused by the formation of hematomas and, ultimately, serious disability.
We used CSD in posterior surgery on the cervical spine. However, we encountered a case in which a large amount of bleeding (drainage) occurred as soon as the suction pressure of the drain was applied immediately after surgery despite minimal intraoperative bleeding. Although the surgical wound was immediately opened and examined, no obvious bleeding point was identified. Therefore, our institution changed the use of CSD to that of closed nonsuction drains (CNSD) in posterior surgery on the cervical spine. The present study investigated and compared the postoperative complications and clinical outcomes of posterior surgery on the cervical spine using the two types of drains (CSD and CNSD).
Materials and Methods
Patients
The present study was performed with approval from the local Institutional Ethics Committee.
Patients who underwent cervical laminoplasty (CLP) without instrumentation were included in the present study. A list of consecutive patients was extracted from the prospectively recorded surgery list between 2014 and 2020, and their clinical data were retrospectively reviewed. We excluded patients with severe liver function failure, hemophilia, platelet abnormalities, active malignancy, and intraoperative dural tears; those who were <20 years old; and those who were followed up for <1 year. When the records of patients were checked, CSD was accidentally used for one patient after the drain was changed to CNSD, and, thus, this patient was also excluded. Therefore, 100 consecutive patients were extracted before and after the drain type was changed (CSD and CNSD groups, respectively), and these two groups were compared.
Primary outcome measures were the drainage amount, blood counts before and after surgery, and postoperative complications, including surgical site infection, the formation of symptomatic hematomas, C5 palsy, and the need for blood transfusion. Fluid collection at the surgical site was evaluated using magnetic resonance images (MRI) taken 10-14 days after surgery. All T2-weighted axial views were examined, and the area of fluid collection adjacent to the epidural space was measured using ImageJ64 software (version 1.46r; NIH, Bethesda, MD). The largest fluid collection areas were categorized as small, medium, and large relative to the area of the canal space of the C2 vertebra (the C2 area). The fluid collection area was defined as small when it was smaller than the C2 area, medium when it was larger than the C2 area and smaller than twice the size of the C2 area, and large when it was larger than twice the size of the C2 area (Fig. 1). To assess disease severity, Japanese Orthopaedic Association (JOA) scores for the cervical spine6) before and 1 year after surgery were evaluated.
Figure 1.
Fluid collection areas were measured using T2-weighted magnetic resonance images. In all axial images, an image of the largest fluid collection area was selected, and the area of fluid correction was compared with that of the canal space at the C2 vertebral body (the C2 area) (A, sagittal view). The fluid collection area was defined as small when the area of fluid collection was smaller than the C2 area (B), medium when the fluid collection area was larger than the C2 area and smaller than twice the size of the C2 area (C), and large when the fluid collection was larger than twice the size of the C2 area (D).
Potential risk factors affecting clinical outcomes, including sex, age, body mass index (BMI), smoking, comorbidities (diabetes, rheumatoid arthritis, dialysis, and hypertension), blood pressure on admission, surgical data, and blood counts, were analyzed.
Blood tests and calculation of total and hidden blood loss
We included patients with blood counts tested within 3 weeks before and 7-10 days after surgery, at which time hemodynamic stability was expected to be achieved7). Patients with blood transfusion or dialysis, or both, were excluded. Therefore, 78 patients in the CSD group and 90 in the CNSD group were analyzed. Hemoglobin levels before and after surgery (Hbpre and Hbpost, respectively) and the hematocrit level before and after surgery (Hctpre and Hctpost, respectively) were recorded. The average of Hctpre and Hctpost was also calculated (Hctave).
Patient blood volume (PBV) was estimated using the formula reported by Nadler8). Total blood loss (TBL) was calculated using the Gross equation9), and hidden blood loss (HBL) was also calculated7).
PBV (mL)=(k1×height (m3) + k2×weight (kg) + k3)×1000
For men, k1=0.3669, k2=0.03219, and k3=0.6041
For women, k1=0.3561, k2=0.03308, and k3=0.1833
TBL (mL)=(PBV×(Hctpre−Hctpost))/Hctave
HBL (mL)=TBL−estimated blood loss at surgery−drainage amount
Surgical procedure for open door laminoplasty
All patients underwent open door laminoplasty as previously reported10). Briefly, the patient's head was fixed with a Mayfield head clamp in the prone position. Using a midline skin incision, the lamina surface was exposed. The center of the lamina was split using a high-speed diamond bar with a diameter of 3 mm. A bony groove was created on the medial side of the facet joint using a diamond bar with a diameter of 5 mm, and the lamina was opened bilaterally. The open lamina was fixed using a suture anchor inserted into the lateral mass11). Partial laminectomy was added to the caudal and cranial sides of the opened laminae in a case-dependent manner. A single drainage tube was placed before wound closure. Before January 2018, an SB tube (Sumitomo Bakelite, Co., Ltd., Tokyo, Japan) was used (CSD, Fig. 2A), which is the same tube type to the Jackson-Pratt drainage tube1). SILASCON (Kaneka Medical, Co., Ltd., Osaka, Japan) has since been employed (CNSD, Fig. 2B) and has same tube type to the Jackson-Pratt drainage tube. Drainage tubes were removed 24-36 hours after surgery in all patients. Patients were allowed to start rehabilitation after drainage tube removal by using a soft cervical orthosis for 2 weeks.
Figure 2.
A, The closed suction drain used before January 2018. B, The closed nonsuction drain used after January 2018.
Statistical analysis
A t-test was used for comparisons between two groups when normality was confirmed, and the Mann-Whitney U test was used for those without normality. Fisher's exact test was employed to analyze the frequencies of variables. Changes in Hb, Hct, and JOA scores were verified using repeated measures analysis of variance. To predict the drainage amount, a multivariable linear regression analysis was used to examine the relationships between the drainage amount and independent variables using the stepwise elimination of variables. A multivariable logistic regression analysis was performed to assess the influence of possible risk factors on the size of hematomas (small or medium and large). The 95% confidence interval of the odds ratio was calculated. A P value <0.05 was considered significant in all analyses. All calculations were performed using R (R for 2.14.1 GUI 1.43).
Results
Demographic and baseline clinical data are shown in Table 1. No significant differences were observed in any items, including age, BMI, smoking, antiplatelet or anticoagulant drug use, comorbidities, and blood pressure, between the two groups. The most common index pathology was cervical spondylotic myelopathy in both groups, accounting for >80% of cases.
Table 1.
Comparison of Demographic and Clinical Data between Patients Using Suction Drain and Those Using Nonsuction Drain.
| CSD group | CNSD group | P | |
|---|---|---|---|
| Number of patients | 100 | 100 | |
| Sex | |||
| Male | 66 | 67 | 1 |
| Female | 34 | 33 | |
| Age±SD (years) | 68.7±11.9 | 69.6±10.7 | 0.56 |
| BMI±SD | 23.9±4.5 | 24.3±4.9 | 0.51 |
| Smoking | 0.94 | ||
| Current | 14 | 12 | |
| Former | 32 | 32 | |
| Never | 54 | 56 | |
| Diabetes | 18 | 23 | 0.38 |
| Hypertension | 43 | 51 | 0.36 |
| Antiplatelet or coagulant drug | 30 | 25 | 0.53 |
| Rheumatoid arthritis | 2 | 6 | 0.17 |
| Dialysis | 4 | 1 | 0.37 |
| Pathology | 0.38 | ||
| CSM | 83 | 89 | |
| OPLL | 8 | 3 | |
| Trauma | 3 | 4 | |
| Others | 6 | 4 | |
| BP–systolic at admission±SD (mmHg) | 131 (20.7) | 135 (18.0) | 0.17 |
| BP–diastolic at admission±SD (mmHg) | 76.2 (11.0) | 76.2 (12.9) | 0.98 |
CSD, closed suction drain; CNSD, closed nonsuction drain; SD, standard deviation; BMI, body mass index; CSM, cervical spondylotic myelopathy; OPLL, ossification of posterior longitudinal ligament; BP, blood pressure
Clinical results of surgery
As shown in Table 2, no significant differences were observed in any surgical parameters, including the number of levels decompressed, surgical times, or estimated blood loss, between the CSD and CNSD groups. Postoperative complications included C5 paralysis in four patients each, SSI in one patient each, and blood transfusion in one patient each in the CSD and CNSD groups; however, no patient developed symptomatic hematoma. The amount of drainage was significantly greater in the CSD group than in the CNSD group (224 vs. 143 mL) (Fig. 3). Epidural fluid collection was evaluated in 70 and 79 patients in the CSD and CNSD groups, respectively. Small and medium to large fluid collection areas were significantly more common in the CSD and CNSD groups, respectively. No significant differences were observed between baseline and 1 year postoperative JOA scores between the two groups.
Table 2.
Comparison of Clinical Data between Patients Using Suction Drain and Those Using Nonsuction Drain.
| CSD group | CNSD group | P value | |
|---|---|---|---|
| Number of patients | 100 | 100 | |
| Number of levels decompressed±SD | 4.4±1.0 | 4.3±1.0 | 0.40 |
| Surgical time±SD (min) | 87.1±28.0 | 84.2±30.1 | 0.50 |
| Estimated blood loss±SD (mL) | 28.3±42.1 | 24.6±45.3 | 0.55 |
| Drainage amount±SD (mL) | 224±115 | 143±74 | <0.001 |
| Complications | |||
| C5 palsy | 4 | 4 | 1.0 |
| Surgical site infection | 1 | 1 | 1.0 |
| Symptomatic hematoma | 0 | 0 | 1.0 |
| Blood transfusion | 1 | 1 | 1.0 |
| JOA score (preoperative) | 11.2 | 11.5 | 0.36† |
| JOA score (1-year postoperative) | 14.4 | 14.3 | <0.0001* |
| Blood test and calculation | N=78 | N=90 | |
| Hbpre±SD (g/dL) | 13.6±1.5 | 13.7±1.5 | 0.0028† |
| Hbpost±SD (g/dL) | 12.2±1.4 | 12.7±1.4 | <0.0001* |
| Hctpre±SD | 40.4±4.3 | 40.5±4.0 | 0.00026† |
| Hctpost±SD | 36.1±3.8 | 37.6±3.9 | <0.0001* |
| Total blood loss±SD (mL) | 338±188 | 245±191 | 0.0024 |
| Hidden blood loss±SD (mL) | 96.1±206 | 80.0±194 | 0.60 |
| Epidural fluid collection | N=70 | N=79 | |
| Small | 29 | 17 | 0.033 |
| Medium | 27 | 40 | |
| Large | 14 | 22 |
CSD, closed suction drain; CNSD, closed nonsuction drain; SD, standard deviation; Hbpre, hemoglobin count before surgery; Hctpre, hematocrit before surgery; Hbpost, hemoglobin count after surgery; Hctpost, hematocrit after surgery; JOA, Japanese Orthopaedic Association; *, Time (preoperative and postoperative) effect; †, Time×Factor effect
Figure 3.
Average value and distribution of drainage volumes between patients with closed nonsuction and suction drains.
Regarding blood counts, Hbpre and Hctpre did not significantly differ between the CSD and CNSD groups; however, Hbpost and Hctpost were significantly decreased in the CSD group than in the CNSD group (Fig. 4). TBL was also significantly greater in the CSD group than in the CNSD group. The difference in TBL between the two groups (93 mL, N=168) was similar to that in the drainage amount between the two groups (81 mL, N=200), and HBL was similar in the CSD and CNSD groups (96.1 vs. 80.0 mL, P=0.60).
Figure 4.
Changes in the hematocrit (upper panel) and hemoglobin levels (lower panel).
Analysis of factors affecting the drainage amount
A multiple regression analysis identified the drain type (t value=6.3) and number of decompressed levels (t value=5.6) as factors affecting the drainage amount (Table 3). BMI and male sex were also risk factors for increases in the drainage amount.
Table 3.
Multiple Regression Analysis to Estimate the Amount of Drainage.
| Variables | Estimates | Std. error | t value | P |
|---|---|---|---|---|
| BMI | 4.3 | 1.5 | 2.8 | 0.0054 |
| Num. decompressed level | 35.3 | 6.3 | 5.6 | <0.001 |
| CSD (vs. CNSD) | 80.1 | 12.7 | 6.3 | <0.001 |
| Male (vs. Female) | 48.6 | 13.6 | 3.6 | <0.001 |
BMI, body mass index; CSD, closed suction drain; CNSD, closed nonsuction drain
Analysis of factors affecting the amount of epidural fluid collection
As shown in Table 4, low systolic blood pressure at admission and the use of CNSD were risk factors for medium or large epidural fluid collection in the univariable analysis. In the multivariable regression analysis, only the use of CNSD was identified as an independent significant risk factor for medium or large epidural fluid collection.
Table 4.
Analysis of Factors Affecting the Amount of Epidural Fluid Collection.
| Univariable analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|
| Small | Medium or Large | P | Odds ratio | 95%CI | P | |
| Number of patients | 46 | 103 | ||||
| Sex | 0.094 | |||||
| Male | 26 (65%) | 73 (72%) | 2.1 | 0.77–5.6 | 0.15 | |
| Female | 20 (35%) | 30 (28%) | Control | |||
| Age±SD (years) | 71.5±10.7 | 69.5±11.3 | 0.31 | Not included | ||
| BMI±SD | 24.3±4.7 | 23.9±3.7 | 0.53 | 0.97 | 0.84–1.1 | 0.52 |
| Smoking | 0.88 | Not included | ||||
| Current | 5 (11%) | 13 (13%) | ||||
| Former | 16 (35%) | 32 (31%) | ||||
| Never | 25 (54%) | 58 (56%) | ||||
| Diabetes | 7 (15%) | 25 (24%) | 0.28 | 1.8 | 0.55–5.8 | 0.33 |
| Hypertension | 22 (48%) | 46 (45%) | 0.73 | 0.72 | 0.28–1.8 | 0.48 |
| Antiplatelet or coagulant drug | 15 (33%) | 27 (26%) | 0.44 | 0.94 | 0.34–2.6 | 0.89 |
| BP–systolic at admission±SD (mmHg) | 138.8±16.9 | 132.9±17.0 | 0.050 | 1.0 | 0.97–1.0 | 0.94 |
| BP–diastolic at admission±SD (mmHg) | 77.9±11.5 | 75.7±11.6 | 0.28 | Not included | ||
| Number of levels decompressed±SD | 4.2±1.1 | 4.4±1.0 | 0.31 | 1.4 | 0.91–2.3 | 0.12 |
| Drainage amount±SD (mL) | 182±131 | 183±91 | 0.99 | Not included | ||
| Estimated blood loss±SD (mL) | 28.8±56.9 | 19.0±26.9 | 0.15 | Not included | ||
| Hbpre±SD (g/dL) | 13.2±1.6 | 13.9±1.5 | 0.044 | 1.3 | 0.95–1.8 | 0.10 |
| Hctpre±SD | 39.5±4.2 | 41.2±4.0 | 0.045 | Not included | ||
| CNSD | 17 (37%) | 62 (60%) | 0.013 | 2.86 | 1.3–6.1 | 0.0061 |
SD, standard deviation; BMI, body mass index; BP, blood pressure; Hbpre, hemoglobin count before surgery; Hctpre, hematocrit before surgery; CNSD, closed nonsuction drain
Discussion
This is the first study to report differences in clinical outcomes between different types of drains, CSD and CNSD, after CLP. One of the strengths of the present study is that MRI was performed on 75% of patients in the early postoperative period, and, thus, the amount of epidural fluid collection was examined. Furthermore, the type of drain was not selected on the basis of the preferences of surgeons, but changed over time. Therefore, the CSD and CNSD groups were assumed to be homogeneous, but differed according to the time of surgery. To ensure that similar surgical techniques were performed, only CLP was included in the present study; patients on whom posterior instrumentation was performed were excluded. No significant differences were observed in the backgrounds of these patients, including demographic data and comorbidities, or surgical factors, such as the number of levels decompressed and estimated blood loss.
The present results showed a mean difference of 81 mL in the postoperative drainage volume between 100 patients in the CSD group and 100 patients in the CNSD group. Although no significant differences were observed in Hbpre and Hctpre between the two groups, both Hbpost and Hctpost were significantly lower in the CSD group than in the CNSD group. TBL calculated from Hct was also higher in the CSD group by an average of 93 mL, which was similar to the difference noted in the postoperative drainage volume, suggesting that the difference in TBL was derived solely from differences in the postoperative drainage volume between the CSD group and CNSD group. HBL did not significantly differ between the two groups, which supports this result. CLP had a short surgical time and low intraoperative blood loss, with only one patient in each group requiring a blood transfusion. Postoperative Hb levels were slightly higher in the CNSD group than in the CSD group, which suggests the superiority of CNSD. In the present study, differences in Hbpost and Hctpost were significant but small and, thus, may not have significant clinical implications. However, in highly invasive surgeries, such as posterior fixation, these differences may increase and affect clinical outcomes. Conversely, postoperative wound fluid collection was significantly larger in the CNSD group than in the CSD group, and the multivariate analysis also identified the use of CNSD as a significant risk factor for medium to large fluid collection. Symptomatic hematomas were not detected in either group; however, because they infrequently occur in CLP, this result needs to be confirmed in a larger study. Regarding infection, one case of infection was found in each group, and this also needs to be verified in a larger case series. Regarding functional outcomes, no significant differences were observed in JOA scores between the two groups before and after surgery, and the type of drainage did not appear to affect clinical outcomes at 1 year after surgery.
The limitation of the present study is its retrospective nature; therefore, the same surgeons did not perform surgery, and there may have been omissions in data, such as comorbidities and medication. Additionally, because the timing of blood collection was not consistent, blood collection data were only examined in 168 cases (84% of all cases). This was also the case for MRI, which was performed on 75% of patients. Another limitation is the small number of cases with complications, such as infection and symptomatic epidural hematoma. We estimated the incidence of each to be approximately 1%, and even if we assume an incidence error of 0.5% with a 95% confidence interval, we need approximately 1500 cases in each group.
The placement of CSD reduces the amount of blood accumulating in a closed wound, which theoretically inhibits the formation of hematomas, wound-healing complications, and infection12). In contrast, the presence of drains may lead to bacterial invasion and increased infection rates with prolonged indwelling13). Limited information is currently available on the use of drains at the cervical or thoracic spine in which the spinal cord resides. A previous study compared the outcomes of 324 patients with and 176 without drains following surgery for adolescent idiopathic scoliosis, which does not generally require decompression, and found no significant differences in infection rates, including early and late infections. Conversely, because the number of cases requiring blood transfusion was higher in the group using drains, it was concluded that the use of drains is a risk factor for blood transfusion14). Regarding posterior surgery on the cervical spine, Herrik et al.15) conducted a retrospective study on 1799 patients and found fewer reoperations for SSI in the group with drains than in the group without drains after adjustments for the presence of diabetes and number of operative levels; however, the reoperation rate for hematomas did not significantly differ between the two groups. These studies were retrospective and did not eliminate selection bias regarding the use of drains.
In the lumbar spine, an RCT on 28 patients each with and without suction drains in minimally invasive transforaminal lumbar interbody fusion reported a shorter hospital stay of 1 day without drains; however, other clinical outcomes did not significantly differ5). Gubin et al.4) performed RCT on the use of drains in posterior lumbar surgery at multiple levels of the lumbar spine; in their analysis of 161 cases, they noted that TBL was greater and significantly more cases required blood transfusion in the drain group. Conversely, postoperative aspiration was significantly more common in the no-drain group; however, no significant differences were observed in wound complications, such as infection.
The lack of RCT on the use of drains in posterior surgery on the cervical spine is due to concerns regarding the worsening of neurological symptoms due to the potential formation of hematomas in the absence of drains. In the present study, the use of CNSD was confirmed to decrease TBL. However, postoperative epidural fluid collection was larger in the CNSD group, and, thus, the risk of hematoma-induced paralysis may be higher in those without drains than in those with drains. The present results suggest that CNSD is useful for decreasing TBL without increasing the risk of complications, such as wound infection and formation of hematomas.
Conclusions
The use of CNSD in CLP has the potential to decrease the postoperative drainage amount and TBL. Although there was no difference in the incidence of perioperative complications such as symptomatic hematoma or surgical site infection, the amount of epidural fluid collected was larger in the CNSD group than in the CSD group.
Conflicts of Interest: The authors declare that there are no relevant conflicts of interest.
Sources of Funding: No funds were received in support of this study.
Author Contributions: BO: Conceptualization, Methodology, Data Curation, and Writing-Original draft preparation. SF, TS, and KM: Visualization, Investigation, and Writing-Reviewing and Editing. SM: Supervision, Reviewing, and Editing.
Ethical Approval: This study was approved by the institutional review board of Kyoto University (R2364).
Informed consent of the patients was not required since this was a retrospective study that used anonymous clinical data. In addition, we used the Opt-out method, which guaranteed the opportunity to refuse to participate in the study after knowing about the conduct and objectives of the study on our website.
References
- 1.Jackson FE, Pratt RA, 3rd. Technical report: a silicone rubber suction drain for drainage of subdural hematomas. Surgery. 1971;70(4):578-9. [PubMed] [Google Scholar]
- 2.Muthu S, Ramakrishnan E, Natarajan KK, et al. Risk-benefit analysis of wound drain usage in spine surgery: a systematic review and meta-analysis with evidence summary. Eur Spine J. 2020;29(9):2111-28. [DOI] [PubMed] [Google Scholar]
- 3.von Eckardstein KL, Dohmes JE, Rohde V. Use of closed suction devices and other drains in spinal surgery: results of an online, germany-wide questionnaire. Eur Spine J. 2016;25(3):708-15. [DOI] [PubMed] [Google Scholar]
- 4.Gubin AV, Prudnikova OG, Subramanyam KN, et al. Role of closed drain after multi-level posterior spinal surgery in adults: a randomised open-label superiority trial. Eur Spine J. 2019;28(1):146-54. [DOI] [PubMed] [Google Scholar]
- 5.Hung PI, Chang MC, Chou PH, et al. Is a drain tube necessary for minimally invasive lumbar spine fusion surgery? Eur Spine J. 2017;26(3):733-7. [DOI] [PubMed] [Google Scholar]
- 6.Yonenobu K, Abumi K, Nagata K, et al. Interobserver and intraobserver reliability of the japanese orthopaedic association scoring system for evaluation of cervical compression myelopathy. Spine. 2001;26(17):1890-4. [DOI] [PubMed] [Google Scholar]
- 7.Sehat KR, Evans RL, Newman JH. Hidden blood loss following hip and knee arthroplasty-correct management of blood loss should take hidden loss into account. J Bone Surg. 2004;86b(4):561-5. [PubMed] [Google Scholar]
- 8.Nadler SB, Hidalgo JH, Bloch T. Prediction of blood volume in normal human adults. Surgery. 1962;51(2):224-32. [PubMed] [Google Scholar]
- 9.Gross JB. Estimating allowable blood loss: corrected for dilution. Anesthesiology. 1983;58(3):277-80. [DOI] [PubMed] [Google Scholar]
- 10.Miyata M, Neo M, Fujibayashi S, et al. Double-door cervical laminoplasty with the use of suture anchors technical note. J Spinal Disord Tech. 2008;21(8):575-8. [DOI] [PubMed] [Google Scholar]
- 11.Kurokawa Y, Yokoyama Y, Kuroda K, et al. Biomechanical evaluation of the suture anchors used in open-door laminoplasty. Spine. 2014;39(21):E1248-55. [DOI] [PubMed] [Google Scholar]
- 12.Aleem IS, Tan LA, Nassr A, et al. Surgical site infection prevention following spine surgery. Global Spine J. 2020;10(1 Suppl):92S-8S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pennington Z, Lubelski D, Molina C, et al. Prolonged post-surgical drain retention increases risk for deep wound infection after spine surgery. World Neurosurg. 2019;130:e846-53. [DOI] [PubMed] [Google Scholar]
- 14.Diab M, Smucny M, Dormans JP, et al. Use and outcomes of wound drain in spinal fusion for adolescent idiopathic scoliosis. Spine. 2012;37(11):966-73. [DOI] [PubMed] [Google Scholar]
- 15.Herrick DB, Tanenbaum JE, Mankarious M, et al. The relationship between surgical site drains and reoperation for wound-related complications following posterior cervical spine surgery: a multicenter retrospective study. J Neurosurg Spine. 2018;29(6):628-34. [DOI] [PubMed] [Google Scholar]