- Email: [email protected]
Prolonged Post-surgical Drain Retention Increases Risk for Deep Wound Infection After Spine Surgery
Original Article
Prolonged Post-surgical Drain Retention Increases Risk for Deep Wound Infection After Spine Surgery Zach Pennington, Daniel Lubelski, Camilo Molina, Erick M. Westbroek, A. Karim Ahmed, Daniel M. Sciubba
OBJECTIVE: Surgical site infections (SSIs) complicate 1% to 9% of elective spine surgeries. Previously identified risk factors include diabetes mellitus type 2, obesity, and chronic kidney disease. We sought to determine whether the use of postoperative surgical site drains is associated with deep SSIs.
-
METHODS: We retrospectively identified patients operated for deep SSIs after surgery for degenerative spine pathologies between July 2016 and December 2018. Patients were excluded if the reason for operation was other than deep SSI or if their primary operation was for infection or tumor. Using their index procedure and the electronic medical record, patients were matched to controls based on age, surgical invasiveness, International Classification of Diseases, Tenth Revision, Clinical Modification code, race, and sex. Our main outcome of interest was whether drain retention time, total output, or daily output differed significantly between cases and controls.
suggest early postoperative drain removal may potentially decrease the risk of SSI and shorten duration of hospital stay.
-
RESULTS: We identified 38 patients who met inclusion criteria. Infected patients had a higher body mass index (34.2 vs. 29.9 kg/m2; P [ 0.001), higher odds of having diabetes mellitus type 2 (55.3% vs. 18.4%; P [ 0.002), longer drain retention time (5.5 vs. 3.5 days; P [ 0.02), and longer inpatient stay (9.5 vs. 4.3 days; P [ 0.005). Multivariable logistic regression demonstrated that even after controlling for the other risk factors, drain retention time independently predicted postoperative surgical site infection (odds ratio: 1.36; P [ 0.02).
-
CONCLUSIONS: Prolonged surgical drain retention correlates with risk of deep SSI after surgery for degenerative spine disease independent of surgical invasiveness, diabetes mellitus type 2 status, and total drain output. Our data
-
Key words Revision surgery - Surgical drain - Surgical site infection - Wound infection -
Abbreviations and Acronyms SSI: Surgical site infection
INTRODUCTION
B
etween 0.72% and 8.7% of patients will experience a surgical site infection (SSI) after elective spine surgery.1,2 SSIS prolong hospital stays,3,4 increase total care 1,3,5 costs, increase patient morbidity,1,4 and worsen patient-reported outcomes.3,6 Prior studies have tried to identify risk factors for surgical site infection, as well as interventions to decrease the rates of SSI. Risk factors include smoking,1,7,8 obesity,1,7-10 invasive surgery,2,8 hypertension,1,7 undergoing a revision versus index surgery,1,2,7 and diabetes mellitus type 2.1,7,8 Some authors have suggested that the use of a surgical site drain may decrease the risk of infection, perhaps by preventing hematoma formation at the wound site.11 Other studies have had conflicting results regarding whether the use of drains or duration of drain placement increases the risk of wound infection.12-15 Nevertheless, a large proportion of patients undergoing surgery for degenerative spine conditions will have a surgical drain placed. These drains can substantially increase the duration of inpatient stays as surgical teams often wait for drain output to drop below a critical threshold prior to discharging the drain and patient.16,17 In the absence of demonstrated benefit with regard to SSI or wound hematoma, excess drain indwelling merely serves to increase care costs.12,13 Consequently, we elected to investigate the correlation of surgical drain usage, retention time, and output with development of SSIs. Our hypothesis was that prolonged surgical drain retention would increase the risk of deep SSI.
To whom correspondence should be addressed: Daniel M. Sciubba, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.013 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
Department of Neurosurgery, Johns Hopkins School of Medicine, Baltimore, Maryland, USA
WORLD NEUROSURGERY -: e1-e8, - 2019
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
METHODS We retrospectively identified all patients who underwent surgical revision for wound infection by the neurosurgical spine service at the Johns Hopkins Hospital between 1 July 2016 and 31 December 2018. Prior to proceeding, institutional review board approval was obtained (IRB00112273). Patients with deep SSIs were identified using the definition for deep wound infection used by the Centers for Disease Control.18 This definition lists deep SSIs as those involving the deep incisional soft tissues, that occur within 90 days of surgery, and are accompanied by one of the following: identification of an abscess or sign of wound infection on imaging or examination, purulent drainage from the wound site, or wound dehiscence or purposeful opening of the incision by a provider with positive wound cultures and evidence of fever or wound tenderness. Included patients were adults aged >18 years having undergone prior surgery for a degenerative condition of the spine less than 3 months prior to readmission (International Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] codes: M40e43, M47, M48.0eM48.2, M48.4eM48.5, M48.8, M49, M50eM51, M53eM54). Patients were excluded if: 1) the indication for the patient’s previous surgery was tumor, trauma, or infection; 2) the patient had a known history of intravenous drug use; or 3) the reason for surgical revision was wound dehiscence without infection (Figure 1). After identification of this case cohort, a second cohort of patients operated over the same time period for degenerative conditions were identified, individually matching the cases based upon surgical indication (ICD-10-CM), age, sex, race, and surgical invasiveness on the adult spinal deformity index (ASD-S), as described by Neuman et al.19 The ASD-S was developed for the adult deformity population and incorporates the number of levels undergoing interbody fusion, laminectomy, pedicle screw instrumentation, and decompression to define overall surgical invasiveness. Prior comparison with the surgical invasiveness index (SII)—a competing measure of assessing surgical invasiveness—has shown the ASD-S to more effectively account for blood loss and operative time, which have themselves been suggested to increase the risk of wound infection.20
Data points gathered for patients included demographics (age, race, gender, body mass index, procedure) and surgical details, including indication for surgery, surgical invasiveness on the ASDS index, surgical levels, index versus revision procedure status, instrumentation, and decompression use. We also gathered data on other endpoints previously established as risk factors for wound infection, including smoking,1,7,8 hypertension,1,7 diabetes mellitus type 2 status,1,7,8 chronic steroid use (defined as 10 days in the month prior to surgery),21 and American Society of Anesthesiologists score.1 To examine the effect of drain usage, we considered the proportion of cases and controls having drain placement. Additionally, among concordant pairs—those in which a drain was placed in both the case and the control—we considered mean drain indwelling time (days), total drainage (mL), average drainage (mL/24 hours), peak drainage (mL/24 hours), and drainage at time of discharge (mL/24 hours). There was no standard protocol regarding when the surgical drain was removed. However, for most patients the drain was removed shortly before discharge. For patients experiencing wound infection, we collected data on the presence or absence of growth on sample culture, the infective agent, and, for cases with Staphylococcus aureus as the principal agent, the presence or absence of methicillin susceptibility. Hospital Infection Protocol At our hospital several steps are taken for all patients undergoing elective spine surgery to reduce the risk of postoperative infection. These include providing patients with 2% chlorhexidine gluconate skin preparation wipes prior to surgery, acquiring nasal swabs for methicillin-resistant S. aureus testing in the preoperation unit, scrubbing the surgical site with chlorhexidine solution prior to incision, administration of intravenous cefazolin within 60 minutes prior to skin incision (2 g for most patients and 3 g for those >120kg; redosing performed every 3e4 hours), and draping of the surgical site with iodine-impregnated drapes. Wound irrigation with vancomycin and continuation of intravenous antibiotics postoperatively are nonstandard practices, and are used variably by some of our surgeons. Similarly, no standard protocol exists for
Figure 1. Schematic illustrating screening of patients readmitting for wound revision.
e2
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
drain placement or discharge timing at our hospital. Nevertheless, most providers elect to place drains in cases using posterior approaches covering 2 or more levels and elect to discontinue their surgical site drain when daily output drops below 50e100 mL per day. Analysis Data were collected using Microsoft Excel (Redmond, WA) and analyzed using TIBCO Statistica 13.3 (Palo Alto, CA). Cases and controls were analyzed as matched pairs. Within each cohort, mean values were reported as mean standard deviation for continuous variables, or as percentages for dichotomous and categorical variables. Inferential univariable statistics were performed using paired t tests for continuous variables, Fisher Exact tests for dichotomous variables, and c2 tests for categorical variables with more than 2 possible outcomes. To define the independently significant variables in those receiving surgical drains, we performed a stepwise analysis, including only variables identified as significant at the P 0.15 level in the multivariable logistic regression. The final regression model was internally validated using 100 bootstrapped samples, with the accuracy being reported as both the C-statistic of the associated receiver operating curve and the mean absolute error across all 100 samples. RESULTS Our query identified 65 patients who underwent surgery for wound revision in the period queried. Of these, 38 cases (58.4%) were performed for wound infection requiring operative washout, which were matched with contemporary controls. As shown in Table 1, cases and controls were similar to one another with regard to sex, race, and surgical invasiveness. There was noted to be a statistically, but not clinically significant, difference in mean age between cases and controls. On univariable analysis (Table 1), cases were found to have significantly greater body mass index (34.2 vs. 29.9 kg/m2; P ¼ 0.001), and were significantly more likely to have diabetes (55.3% vs. 18.4%; P ¼ 0.002), hypertension (81.6% vs. 57.9%; P ¼ 0.04), to have undergone a prior spine surgery (52.6% vs. 21.1%; P ¼ 0.03), and had a longer inpatient stay (9.5 vs. 4.3 days; P ¼ 0.005). A total of 34 of the case-control pairs had drains placed in both the case and the control. In these patients, those experiencing a wound infection were noted to have significantly longer drain indwelling times (5.5 vs. 3.5 days; P ¼ 0.02). Furthermore, a larger proportion of patients experiencing wound infection had their drain removed 1 week or longer after surgery (30.3% vs. 2.9%; P ¼ 0.002). We found no significant difference between cases and controls in terms of total drainage, average drainage, or drainage on the day of drain removal. Lastly, we found no difference in the percentage of drains removed with output greater than 100 mL or 200 mL in the prior 24 hours. On multivariable analysis of the 34 concordant pairs (Table 2), total drain indwelling time (odds ratio ¼ 1.36 per day; P ¼ 0.02) was a significant and independent predictor of postoperative wound infection requiring reoperation. Diabetes mellitus approached but did not reach statistical significance (P ¼ 0.16). Of note, lower preoperative hemoglobin level also predicted
WORLD NEUROSURGERY -: e1-e8, - 2019
DRAIN TIME AND SPINE SURGERY
development of postoperative wound infection (odds ratio ¼ 0.53 per g/dL increase in Hgb; P ¼ 0.005). None of the other variables found to be significant on univariable analysis proved significant on the multivariable analysis. A receiver operating curve constructed using these variables an associated C-statistic of 0.85 (Figure 2). Of infected patients (Table 3), the mean time between index surgery and readmission was 25.3 11.5 days (range: 8e53 days). A total of 33 patients (86.8%) had cultures grow out on their surgical specimens. The most commonly identified infecting agents were S aureus (33.3%), coagulase-negative Staphylococcus (15.2%), and polymicrobial gram negative cultures (12.1%). Of patients infected with S aureus, 4 grew out methicillin-resistant cultures (36.4%). DISCUSSION SSIs are debilitating postoperative complications that increase readmission rates,22,23 prolong hospital stays,3,4 increase overall morbidity,1,4 and produce worse patient outcomes.3,6 SSIs requiring surgical revision also increase direct care costs, with previous estimates varying between $15,000 and $38,000 depending upon the size of the index procedure.1,6,24 Because of this, there is great emphasis placed upon interventions designed to decrease surgical site infection, including placement of vancomycin powder in the surgical site,25 use of preoperative chlorhexidine wipes,26 and intravenous administration of antibiotics during the immediate pre-incisional period.27 Here, our results suggest that expediting surgical drain discharge may be another potential intervention as we found quicker surgical drain discharge was associated with a lower prevalence of deep SSI. A patient’s cumulative risk of surgical site infection is created by the combination of 2 categories of risk factors: 1) fixed factors, such as a patient’s medical morbidities (smoking, body mass index, diabetes mellitus type 2) and 2) modifiable factors, which include surgery duration, surgical technique, and perioperative and postoperative management. The latter category is that which is targeted by efforts to reduce infection risk and it includes the usage and timing of drain removal. There is a paucity of evidence regarding the degree to which surgical drain usage affects the incidence of spinal surgical site infection. Investigations in the orthopedic surgery literature have shown that prolonged surgical drains increase the risk of surgical site infection.28 This is speculated to stem from retrograde transmission of bacteria along the drain tract.29 Some small neurosurgical studies have not been able to replicate the link between drain usage and surgical site infection.12,13 Based on these studies, it is unclear whether drain indwelling time, output thresholds for drain removal, and average drain output affect the risk of SSI. Our data herein suggest that prolonged surgical drain indwelling time increases the risk for surgical site infection, independent of body habitus, surgical invasiveness, and medical history of diabetes mellitus. To our knowledge, only 2 prior studies have examined drain output and/or indwelling time in patients undergoing surgery for degenerative spine disease. Macki et al. examined a series of patients undergoing surgery for degenerative conditions of the
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
Table 1. Demographics and Comparison of Patients Requiring Surgical Revision for Surgical Site Infection Infection
No Infection
38
38
59.9 12.6
62.3 11.5
Male
28.9
39.5
Female
71.1
60.5
Variable n Age, years Sex
Race
Table 1. Continued Infection
No Infection
P Value
No drain
13.2
7.9
—
Variable P Value
Drain d/c POD1-2
10.5
18.4
—
0.02
Drain d/c POD3-4
31.6
52.6
—
0.33
Drain d/c POD 5e7
23.7
21.1
—
Drain d/c after POD 7
21.1
7.9
—
Drain d/c 7d
30.3
2.9
0.002
0.53
Drain d/c < 5d
51.5
77.1
0.02
1241.5 1102.9
912.5 672.2
0.11
Peak drainage, mL/24 hours
359.7 287.1
381.3 244.8
0.68
Average drainage, mL/24 hours
179.5 123.1
197.2 127.3
0.51
64.5 68.9
65.5 86.5
0.96
Drainage on last day 200 mL
24.2
17.1
0.34
Drainage on last day 100 mL
24.2
17.1
0.34
Drainage on last day 50 mL
42.4
37.1
0.42
9.5 10.9
4.3 2.4
0.005
42.4
45.7
0.49
White
71.1
78.9
Total drainage, mL
Black
21.1
18.4
Asian
2.6
2.6
Other
5.3
0
BMI, kg/m2
34.2 8.3
29.9 4.5
0.001
55.3
18.4
0.002
15.8
7.9
0.48
Medical history Diabetes Insulin-dependent Smoking status
0.45
Current
23.7
13.2
Former
23.7
31.6
52.6
55.3
Chronic kidney disease
Never
10.5
5.3
0.67
Congestive heart failure
5.3
5.3
1.0
COPD
15.8
5.3
0.26
HTN
81.6
57.9
0.04
Cancer*
7.9
5.3
1.0
Chronic steroid use
7.9
0
0.24
0
0
1.0
Coagulopathy
Length of stay, days
Surgery data Index surgery
0.03
Index
47.4
78.9
Revision
52.6
21.1
ASA score
0.03
1e2
13.2
36.8
3e4
86.8
63.2
19.3 14.1
18.3 13.6
0.14
278.1 112.9
280.6 128.8
0.86
86.8
92.1
0.71
5.49 4.83
3.51 1.34
0.02
ASD-S score Surgery duration, minutes Postoperative course Drain placed Drain retention time, days
Continues
e4
www.SCIENCEDIRECT.com
Drainage on last day, mL/24 hours
Postop abx given
Values are presented as mean standard deviation or %. Abx, antibiotics; ASA, American Society of Anesthesiologists; ASD-S, adult spinal deformity index; BMI, body mass index; BUN, blood urea nitrogen; COPD, chronic obstructive pulmonary disease; d/c, discharge; HTN, hypertension; POD, postoperative day. *Active malignancy in previous 5 years.
lumbar spine and found that neither drain removal setting (outpatient vs. inpatient) nor drain indwelling time were correlated with the rate of surgical revision for wound infection.30 However, among their cohort it was noted that those with drain retention time in the top quintile had an odds of surgical wound revision more than 6-fold higher than those with the median drain indwelling time, even after accounting for a history of diabetes mellitus type 2 and other comorbidities. Accordingly, it is possible that their lack of significant findings may be related to their small sample size. Herrick et al. examined a multiinstitutional series of patients undergoing posterior cervical spine surgery.31 They noted that although drain placement was an independent, negative predictor of return to the operating room, daily and total drain output were greater in the group of patients requiring reoperation. Their data further showed that patients requiring revision operation had higher drain indwelling time (3.58 vs. 2.34 days). Similar to the study by Macki et al., the authors found no significant relationship between drain indwelling time and the need for reoperation for wound
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
Table 2. Multivariable Logistic Compression Predicting Patients Requiring Surgical Revision for Wound Infection Among Those Being Treated for Degenerative Spine Pathologies with Postoperative Drain Placement 95% CI Predictor
OR
Lower
Upper
P Value
BMI (per kg/m2)
1.18
1.06
1.33
0.004
Hgb (per g/dL)
0.53
0.35
0.82
0.005
Drain retention time (per day)
1.36
1.04
1.78
0.02
ASA, American Society of Anesthesiologists; BMI, body mass index; CI, confidence interval; Hgb, hemoglobin; OR, odds ratio.
infection. Given that infected patients had drains maintained for an average of 1.2 days longer than non-infected patients, it can be suggested that there may have been a trend that just did not reach statistical significance. Given this limited study of the linkage between drain indwelling time and infection risk, the manner in which prolonged drain indwelling increases the risk of surgical site infection is unknown. However, evidence from both animal experiments32,33 and clinical orthopedic literature has suggested that placement of a surgical
Figure 2. Receiver-operating curve demonstrating the accuracy of the multivariable logistic regression
WORLD NEUROSURGERY -: e1-e8, - 2019
drain may enable the tracking of skin flora and other epidermal bacteria into the surgical wound.34 Using rabbit models, both Raves et al.32 and Cerise et al.33 found that placement of closed suction drains (Jackson-Pratt drains) led to significantly higher rates of deep wound infection. Additionally, Cerise et al. found drain placement to be associated with a significant increase in deep drain site abscess formation among 533 patients undergoing splenectomy. Sørensen and Sørensen documented that among 500 orthopedic patients, positive drain tip cultures was associated with an increased risk of infection, as was a drain indwelling time greater than 6 days.34 Similarly, research from the field of reconstructive surgery has found bacterial colonization of surgical drains to increase substantially with prolonged drain indwelling time, with 1 study finding that drain colonization rates tripled from postoperative day 7 to day 14.35 More recently, similar results have been reported in breast surgery, revealing that each additional week of drain indwelling time increases the risk of surgical site infection by 76.2%.36 In aggregate these studies suggest that early drain removal should decrease wound site penetration by the skin microfauna, lowering the risk of deep SSI and commensurately lowering the likelihood of requiring surgical wound revision.37 Despite these potential negative sequelae of surgical drains, other prior studies have found drain use to be associated with a significantly lower rate of wound infection among both degenerative and non-degenerative spine patients.31,38 Additionally, many
incorporating drain indwelling time as an independent predictor of surgical site infection requiring reoperation.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
Table 3. Results of Sample Cultures for Infected Patients Case
Time to Infection
Culture Result
Gram Stain
Infecting Agent
Length of Stay (days)
1
22
Pos
Pos
MRSA
59
2
31
Neg
N/A
Unknown
6
3
17
Pos
Neg
Polymicrobial gram negative
25
4
10
Pos
Pos
Coagulase-negative Staphylococcus
3
5
24
Neg
N/A
Unknown
23
6
42
Pos
Neg
B. thetaiotaomicron
32
7
20
Pos
Neg
E. coli
15
8
22
Pos
Pos
E. faecalis
5
9
15
Pos
Pos
P. acnes
7
10
19
Pos
Pos
MSSA
23
11
26
Pos
Neg
Polymicrobial gram negative
10
12
33
Pos
Pos
MSSA
7
13
46
Pos
Pos
E. faecalis
13
14
36
Pos
Pos
MRSA
7
15
15
Pos
Pos
MRSA
9
16
20
Pos
Neg
E. coli
5
17
21
Pos
Pos
E. faecalis, S. marcescens
19
18
29
Pos
Neg
P. mirabilis
6
19
15
Neg
N/A
Unknown
36
20
21
Pos
Pos
Coagulase-negative Staphylococcus
29
21
34
Pos
Neg
P. mirabilis
7
22
36
Pos
Pos
MSSA
26
23
30
Pos
Pos
Coagulase-negative Staphylococcus
6
24
14
Pos
Pos
MSSA
6
25
31
Pos
Pos
Coagulase-negative Staphylococcus
9
26
39
Pos
Pos
MSSA
7
27
33
Pos
Pos
MSSA
5
28
13
Pos
Pos
S. pneumoniae
31
29
17
Neg
N/A
Unknown
4
30
53
Pos
Pos
MRSA
13
31
15
Pos
Pos
MSSA
2
32
19
Pos
Neg
Polymicrobial gram negative
17
33
14
Pos
Neg
E. coli
11
34
11
Pos
Neg
Polymicrobial gram negative
9
35
47
Pos
Pos
Group B Streptococcus
3
36
20
Pos
Neg
S. marcescens
7
37
43
Pos
Pos
Coagulase-negative Staphylococcus
7
38
8
Neg
N/A
Unknown
11
MRSA, methicillin-resistant staphylococcus aureus; MSSA, methicillin-sensitive staphylococcus aureus; N/A, not available; Neg, negative; Pos, positive.
e6
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
surgeons elect to place a drain for reasons other than infection prophylaxis, including the potential to decrease the risk of wound site hematoma and subsequent neurological compromise.31 The bulk of the evidence has failed to support the utility of drains for this purpose, however,39 although many groups still advocate for removal of the drain only after output drops below 50 mL per day.40-42 Given the lack of evidence in the literature and our observation of no correlation between SSI and final day drain output up to 200 mL per 24-hour period, we believe that for the average patient undergoing surgery for degenerative spine disease, drains may safely be removed while still outputting as much as 200 mL per day. In addition to potentially reducing the risk of SSI, early drain discharge would likely decrease total care costs. The time required for drain output to drop from 100 mL/24 hours to 50 mL/24 hours may prolong hospital stays by an additional 1 or 2 days, which according to the Kaiser Foundation translates into a cost increase of $1,889 to $4,976.43 Assuming the lower estimate and extrapolating to the entire US spinal surgery population (estimated at 500,000 people annually44) we find that total savings to the US health system could approach $500 million to $1 billion. Limitations There are several limitations to this study, including the small sample size. Our study examined only patients treated for the primary indication of SSI after surgery for degenerative conditions of the spine. Small sample populations are more susceptible to bias by outlier observations. In the present cohort one such bias may stem from the inconsistent application of vancomycin powder to the operative wound across surgeons. The proportion of patients receiving this intervention, along with other occult confounders, potentially limit the generalizability of the conclusions that can be drawn from them. Because of this, it
2. Smith JS, Shaffrey CI, Sansur CA, et al. Rates of infection after spine surgery based on 108,419 procedures: a report from the Scoliosis Research Society Morbidity and Mortality Committee. Spine (Phila Pa 1976). 2011;36:556-563. 3. Badia JM, Casey AL, Petrosillo N, Hudson PM, Mitchell SA, Crosby C. Impact of surgical site infection on healthcare costs and patient outcomes: a systematic review in six European countries. J Hosp Infect. 2017;96:1-15. 4. Casper DS, Zmistowski B, Hollern DA, et al. The effect of postoperative spinal infections on patient mortality. Spine (Phila Pa 1976). 2018;43:223-227. 5. Berríos-Torres SI, Umscheid CA, Bratzler DW, et al. Centers for Disease Control and Prevention Guideline for the prevention of surgical site infection, 2017. JAMA Surg. 2017;152:784. 6. Kuhns BD, Lubelski D, Alvin MD, et al. Cost and quality of life outcome analysis of postoperative
CONCLUSIONS Longer drain indwelling time seems to be an independent risk factor for SSIs after surgery for degenerative spine pathologies. Most importantly, drain retention for greater than 1 week appears to be associated with a higher rate of SSIs. To our knowledge, this is the first study to directly correlate drain indwelling time with the occurrence of spinal wound site infections. The results of the present study suggest that drain discontinuation may be considered earlier to reduce infections and decrease length of hospital stay. Using this higher threshold may decrease drain indwelling times and decrease the rate of SSIs among spine surgery patients. Additionally, simple cost analysis suggests that expediting drain discharge may decrease overall care costs within this population.
infections after subaxial dorsal cervical fusions. J Neurosurg Spine. 2015;22:381-386.
REFERENCES 1. Anderson PA, Savage JW, Vaccaro AR, et al. Prevention of surgical site infection in spine surgery. Neurosurgery. 2017;80(3 Suppl):S114-S123.
is possible that our results may not be applicable to all surgical patients. Patients had skin preparation with either chlorhexidinebased or povidone-iodineebased agents as well as iodophoreimpregnated drapes. The evidence is equivocal regarding the superiority of one preparation type versus another and this may potentially lead to variability in incision site colonization and possible infection risk. The small sample size may also reduce the sensitivity of our analysis by hampering our ability to identify true risk factors that have relatively small contributions to a patient’s cumulative risk. Additionally, our study is retrospective in nature, meaning that we are only able to correlate drain indwelling time with the odds of having experienced an SSI; we are unable to determine whether this relationship is causal. In future studies, we hope to use a larger patient cohort from multiple institutions to provide more robust results. Lastly, given the high proportion of patients in both cohorts who received a surgical drain, our study may have had insufficient power to identify the contribution of drain placement versus nonplacement to the occurrence of deep SSI.
7. Meng F, Cao J, Meng X. Risk factors for surgical site infections following spinal surgery. J Clin Neurosci. 2015;22:1862-1866. 8. Yao R, Zhou H, Choma TJ, Kwon BK, Street J. Surgical site infection in spine surgery: who is at risk? Global Spine J. 2018;8(4 suppl):5S-30S. 9. Mehta AI, Babu R, Sharma R, et al. Thickness of subcutaneous fat as a risk factor for infection in cervical spine fusion surgery. J Bone Joint Surg Am. 2013;95:323-328. 10. Mehta AI, Babu R, Karikari IO, et al. 2012 Young Investigator Award winner: the distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine (Phila Pa 1976). 2012;37:1652-1656. 11. Chimenti P, Molinari R. Post-operative spinal epidural hematoma causing American Spinal Injury Association B spinal cord injury in patients with suction wound drains. J Spinal Cord Med. 2013; 36:213-219.
WORLD NEUROSURGERY -: e1-e8, - 2019
12. Andrew Glennie R, Dea N, Street JT. Dressings and drains in posterior spine surgery and their effect on wound complications. J Clin Neurosci. 2015;22:1081-1087. 13. Liu Y, Li Y, Miao J. Wound drains in posterior spinal surgery: a meta-analysis. J Orthop Surg Res. 2016;11:16. 14. Chen Z, Gao Y, Chen W, Li X, Zhang Y. Is wound drainage necessary in hip arthroplasty? A metaanalysis of randomized controlled trials. Eur J Orthop Surg Traumatol. 2014;24:939-946. 15. Zhang Q, Guo W, Zhang Q, Liu Z, Cheng L, Li Z. Comparison between closed suction drainage and nondrainage in total knee arthroplasty: a metaanalysis. J Arthroplasty. 2011;26:1265-1272. 16. Adogwa O, Elsamadicy AA, Sergesketter AR, et al. Post-operative drain use in patients undergoing decompression and fusion: incidence of complications and symptomatic hematoma. J Spine Surg. 2018;4:220-226. 17. Adogwa O, Khalid SI, Elsamadicy AA, et al. The use of subfascial drains after multi-level anterior cervical discectomy and fusion: does the data support its use? J Spine Surg. 2018;4:227-232.
www.journals.elsevier.com/world-neurosurgery
e7
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
18. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Am J Infect Control. 1999;27: 97-134. 19. Neuman BJ, Ailon T, Scheer JK, et al. Development and validation of a novel adult spinal deformity surgical invasiveness score: analysis of 464 patients. Neurosurgery. 2018;82:847-853. 20. Pawar AY, Biswas SK. Postoperative spine infections. Asian Spine J. 2016;10:176-183. 21. Ranson WA, White SJW, Cheung ZB, et al. The effects of chronic preoperative steroid therapy on perioperative complications following elective posterior lumbar fusion. Glob Spine J. 2018;8: 834-841. 22. Lee JJ, An SB, Kim TW, et al. Analysis of risk factors associated with hospital readmission within 360 days after degenerative lumbar spine surgery in elderly patients. World Neurosurg. 2019: e196-e207. 23. Park P, Nerenz DR, Aleem IS, et al. Risk factors associated with 90-day readmissions after degenerative lumbar fusion: an examination of the Michigan Spine Surgery Improvement Collaborative (MSSIC) Registry. Neurosurgery. 2018:1-7. 24. McGirt MJ, Parker SL, Lerner J, Engelhart L, Knight T, Wang MY. Comparative analysis of perioperative surgical site infection after minimally invasive versus open posterior/transforaminal lumbar interbody fusion: analysis of hospital billing and discharge data from 5170 patients. J Neurosurg Spine. 2011;14:771-778. 25. He X, Sun T, Wang J, Li G, Fei Q. Application of vancomycin powder to reduce surgical infection and deep surgical infection in spinal surgery: a meta-analysis. Clin Spine Surg. 2019;32:150-163. 26. Savage JW, Anderson PA. An update on modifiable factors to reduce the risk of surgical site infections. Spine J. 2013;13:1017-1029. 27. Barker FG II. Efficacy of prophylactic antibiotic therapy in spinal surgery: a meta-analysis. Neurosurgery. 2002;51:391-401. 28. Saleh K, Olson M, Resig S, et al. Predictors of wound infection in hip and knee joint
e8
www.SCIENCEDIRECT.com
DRAIN TIME AND SPINE SURGERY
replacement: results from a 20 year surveillance program. J Orthop Res. 2002;20:506-515. 29. Reiffel AJ, Barie PS, Spector JA. A multi-disciplinary review of the potential association between closed-suction drains and surgical site infection. Surg Infect (Larchmt). 2013;14:244-269.
39. Wang L, Liu X. Cervical sagittal alignment in adolescent idiopathic scoliosis patients (Lenke type 1-6). J Orthop Sci. 2017;22:254-259. 40. Kanayama M, Oha F, Togawa D, Shigenobu K, Hashimoto T. Is closed-suction drainage necessary for single-level lumbar decompression?: review of 560 cases. Clin Orthop Relat Res. 2010;468: 2690-2694.
30. Macki M, Uzosike A, Kerezoudis P, Bydon A, Bydon M, Gokaslan ZL. Duration of indwelling drain following instrumented posterolateral fusion of the lumbar spine does not predict surgical site infection requiring reoperation. J Clin Neurosci. 2017;40:44-48.
41. Diab M, Smucny M, Dormans JP, et al. Use and outcomes of wound drain in spinal fusion for adolescent idiopathic scoliosis. Spine (Phila Pa 1976). 2012;37:966-973.
31. 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;29:1-7.
42. Kamath VHD, Cheung JPY, Mak KC, et al. Antimicrobial prophylaxis to prevent surgical site infection in adolescent idiopathic scoliosis patients undergoing posterior spinal fusion: 2 doses versus antibiotics till drain removal. Eur Spine J. 2016;25:3242-3248.
32. Raves JJ, Slifkin M, Diamond DL. A bacteriologic study comparing closed suction and simple conduit drainage. Am J Surg. 1984;148:618-620.
43. Henry J. Kaiser Family Foundation. Hospital adjusted expenses per inpatient day by ownership. Available at: https://www.kff.org/health-costs/ state-indicator/expenses-per-inpatient-day-by-own ership/?currentTimeframe¼0&sortModel¼%7B% 22colId%22:%22Location%22,%22sort%22:%22asc% 22%7D; 2016. Accessed February 17, 2019.
33. Cerise EJ, Pierce WA, Diamond DL. Abdominal drains: their role as a source of infection following splenectomy. Ann Surg. 1970;171:764-769. 34. Sørensen AI, Sørensen TS. Bacterial growth on suction drain tips. Prospective study of 489 clean orthopedic operations. Acta Orthop Scand. 1991;62: 451-454. 35. Felippe WAB, Werneck GL, Santoro-Lopes G. Surgical site infection among women discharged with a drain in situ after breast cancer surgery. World J Surg. 2007;31:2293-2299. 36. Chen C-F, Lin S-F, Hung C-F, Chou P. Risk of infection is associated more with drain duration than daily drainage volume in prosthesis-based breast reconstruction: a cohort study. Medicine (Baltimore). 2016;95:e5605. 37. Kobayashi K, Imagama S, Ito Z, et al. Is a drain tip culture required after spinal surgery? Clin Spine Surg. 2017;30:356-359. 38. Ramo BA, Roberts DW, Tuason D, et al. Surgical site infections after posterior spinal fusion for neuromuscular scoliosis. J Bone Joint Surg Am. 2014; 96:2038-2048.
44. Steiner CA, Karaca Z, Moore BJ, Imshaug MC, Pickens G. Surgeries in Hospital-Based Ambulatory Surgery and Hospital Inpatient Settings, 2014. Rockville, MD: Agency for Healthcare Research and Quality; 2014. Conflict of interest statement: C. Molina is a consultant for Augmedics. D.M. Sciubba is a consultant for Baxter, DePuySynthes, Globus, K2M, Medtronic, NuVasive, and Stryker. The remaining authors have no conflicts to report. Received 29 April 2019; accepted 1 July 2019 Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.013 Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
Prolonged Post-surgical Drain Retention Increases Risk for Deep Wound Infection After Spine Surgery Zach Pennington, Daniel Lubelski, Camilo Molina, Erick M. Westbroek, A. Karim Ahmed, Daniel M. Sciubba
OBJECTIVE: Surgical site infections (SSIs) complicate 1% to 9% of elective spine surgeries. Previously identified risk factors include diabetes mellitus type 2, obesity, and chronic kidney disease. We sought to determine whether the use of postoperative surgical site drains is associated with deep SSIs.
-
METHODS: We retrospectively identified patients operated for deep SSIs after surgery for degenerative spine pathologies between July 2016 and December 2018. Patients were excluded if the reason for operation was other than deep SSI or if their primary operation was for infection or tumor. Using their index procedure and the electronic medical record, patients were matched to controls based on age, surgical invasiveness, International Classification of Diseases, Tenth Revision, Clinical Modification code, race, and sex. Our main outcome of interest was whether drain retention time, total output, or daily output differed significantly between cases and controls.
suggest early postoperative drain removal may potentially decrease the risk of SSI and shorten duration of hospital stay.
-
RESULTS: We identified 38 patients who met inclusion criteria. Infected patients had a higher body mass index (34.2 vs. 29.9 kg/m2; P [ 0.001), higher odds of having diabetes mellitus type 2 (55.3% vs. 18.4%; P [ 0.002), longer drain retention time (5.5 vs. 3.5 days; P [ 0.02), and longer inpatient stay (9.5 vs. 4.3 days; P [ 0.005). Multivariable logistic regression demonstrated that even after controlling for the other risk factors, drain retention time independently predicted postoperative surgical site infection (odds ratio: 1.36; P [ 0.02).
-
CONCLUSIONS: Prolonged surgical drain retention correlates with risk of deep SSI after surgery for degenerative spine disease independent of surgical invasiveness, diabetes mellitus type 2 status, and total drain output. Our data
-
Key words Revision surgery - Surgical drain - Surgical site infection - Wound infection -
Abbreviations and Acronyms SSI: Surgical site infection
INTRODUCTION
B
etween 0.72% and 8.7% of patients will experience a surgical site infection (SSI) after elective spine surgery.1,2 SSIS prolong hospital stays,3,4 increase total care 1,3,5 costs, increase patient morbidity,1,4 and worsen patient-reported outcomes.3,6 Prior studies have tried to identify risk factors for surgical site infection, as well as interventions to decrease the rates of SSI. Risk factors include smoking,1,7,8 obesity,1,7-10 invasive surgery,2,8 hypertension,1,7 undergoing a revision versus index surgery,1,2,7 and diabetes mellitus type 2.1,7,8 Some authors have suggested that the use of a surgical site drain may decrease the risk of infection, perhaps by preventing hematoma formation at the wound site.11 Other studies have had conflicting results regarding whether the use of drains or duration of drain placement increases the risk of wound infection.12-15 Nevertheless, a large proportion of patients undergoing surgery for degenerative spine conditions will have a surgical drain placed. These drains can substantially increase the duration of inpatient stays as surgical teams often wait for drain output to drop below a critical threshold prior to discharging the drain and patient.16,17 In the absence of demonstrated benefit with regard to SSI or wound hematoma, excess drain indwelling merely serves to increase care costs.12,13 Consequently, we elected to investigate the correlation of surgical drain usage, retention time, and output with development of SSIs. Our hypothesis was that prolonged surgical drain retention would increase the risk of deep SSI.
To whom correspondence should be addressed: Daniel M. Sciubba, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.013 Journal homepage: www.journals.elsevier.com/world-neurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
Department of Neurosurgery, Johns Hopkins School of Medicine, Baltimore, Maryland, USA
WORLD NEUROSURGERY -: e1-e8, - 2019
www.journals.elsevier.com/world-neurosurgery
e1
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
METHODS We retrospectively identified all patients who underwent surgical revision for wound infection by the neurosurgical spine service at the Johns Hopkins Hospital between 1 July 2016 and 31 December 2018. Prior to proceeding, institutional review board approval was obtained (IRB00112273). Patients with deep SSIs were identified using the definition for deep wound infection used by the Centers for Disease Control.18 This definition lists deep SSIs as those involving the deep incisional soft tissues, that occur within 90 days of surgery, and are accompanied by one of the following: identification of an abscess or sign of wound infection on imaging or examination, purulent drainage from the wound site, or wound dehiscence or purposeful opening of the incision by a provider with positive wound cultures and evidence of fever or wound tenderness. Included patients were adults aged >18 years having undergone prior surgery for a degenerative condition of the spine less than 3 months prior to readmission (International Classification of Diseases, Tenth Revision, Clinical Modification [ICD-10-CM] codes: M40e43, M47, M48.0eM48.2, M48.4eM48.5, M48.8, M49, M50eM51, M53eM54). Patients were excluded if: 1) the indication for the patient’s previous surgery was tumor, trauma, or infection; 2) the patient had a known history of intravenous drug use; or 3) the reason for surgical revision was wound dehiscence without infection (Figure 1). After identification of this case cohort, a second cohort of patients operated over the same time period for degenerative conditions were identified, individually matching the cases based upon surgical indication (ICD-10-CM), age, sex, race, and surgical invasiveness on the adult spinal deformity index (ASD-S), as described by Neuman et al.19 The ASD-S was developed for the adult deformity population and incorporates the number of levels undergoing interbody fusion, laminectomy, pedicle screw instrumentation, and decompression to define overall surgical invasiveness. Prior comparison with the surgical invasiveness index (SII)—a competing measure of assessing surgical invasiveness—has shown the ASD-S to more effectively account for blood loss and operative time, which have themselves been suggested to increase the risk of wound infection.20
Data points gathered for patients included demographics (age, race, gender, body mass index, procedure) and surgical details, including indication for surgery, surgical invasiveness on the ASDS index, surgical levels, index versus revision procedure status, instrumentation, and decompression use. We also gathered data on other endpoints previously established as risk factors for wound infection, including smoking,1,7,8 hypertension,1,7 diabetes mellitus type 2 status,1,7,8 chronic steroid use (defined as 10 days in the month prior to surgery),21 and American Society of Anesthesiologists score.1 To examine the effect of drain usage, we considered the proportion of cases and controls having drain placement. Additionally, among concordant pairs—those in which a drain was placed in both the case and the control—we considered mean drain indwelling time (days), total drainage (mL), average drainage (mL/24 hours), peak drainage (mL/24 hours), and drainage at time of discharge (mL/24 hours). There was no standard protocol regarding when the surgical drain was removed. However, for most patients the drain was removed shortly before discharge. For patients experiencing wound infection, we collected data on the presence or absence of growth on sample culture, the infective agent, and, for cases with Staphylococcus aureus as the principal agent, the presence or absence of methicillin susceptibility. Hospital Infection Protocol At our hospital several steps are taken for all patients undergoing elective spine surgery to reduce the risk of postoperative infection. These include providing patients with 2% chlorhexidine gluconate skin preparation wipes prior to surgery, acquiring nasal swabs for methicillin-resistant S. aureus testing in the preoperation unit, scrubbing the surgical site with chlorhexidine solution prior to incision, administration of intravenous cefazolin within 60 minutes prior to skin incision (2 g for most patients and 3 g for those >120kg; redosing performed every 3e4 hours), and draping of the surgical site with iodine-impregnated drapes. Wound irrigation with vancomycin and continuation of intravenous antibiotics postoperatively are nonstandard practices, and are used variably by some of our surgeons. Similarly, no standard protocol exists for
Figure 1. Schematic illustrating screening of patients readmitting for wound revision.
e2
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
drain placement or discharge timing at our hospital. Nevertheless, most providers elect to place drains in cases using posterior approaches covering 2 or more levels and elect to discontinue their surgical site drain when daily output drops below 50e100 mL per day. Analysis Data were collected using Microsoft Excel (Redmond, WA) and analyzed using TIBCO Statistica 13.3 (Palo Alto, CA). Cases and controls were analyzed as matched pairs. Within each cohort, mean values were reported as mean standard deviation for continuous variables, or as percentages for dichotomous and categorical variables. Inferential univariable statistics were performed using paired t tests for continuous variables, Fisher Exact tests for dichotomous variables, and c2 tests for categorical variables with more than 2 possible outcomes. To define the independently significant variables in those receiving surgical drains, we performed a stepwise analysis, including only variables identified as significant at the P 0.15 level in the multivariable logistic regression. The final regression model was internally validated using 100 bootstrapped samples, with the accuracy being reported as both the C-statistic of the associated receiver operating curve and the mean absolute error across all 100 samples. RESULTS Our query identified 65 patients who underwent surgery for wound revision in the period queried. Of these, 38 cases (58.4%) were performed for wound infection requiring operative washout, which were matched with contemporary controls. As shown in Table 1, cases and controls were similar to one another with regard to sex, race, and surgical invasiveness. There was noted to be a statistically, but not clinically significant, difference in mean age between cases and controls. On univariable analysis (Table 1), cases were found to have significantly greater body mass index (34.2 vs. 29.9 kg/m2; P ¼ 0.001), and were significantly more likely to have diabetes (55.3% vs. 18.4%; P ¼ 0.002), hypertension (81.6% vs. 57.9%; P ¼ 0.04), to have undergone a prior spine surgery (52.6% vs. 21.1%; P ¼ 0.03), and had a longer inpatient stay (9.5 vs. 4.3 days; P ¼ 0.005). A total of 34 of the case-control pairs had drains placed in both the case and the control. In these patients, those experiencing a wound infection were noted to have significantly longer drain indwelling times (5.5 vs. 3.5 days; P ¼ 0.02). Furthermore, a larger proportion of patients experiencing wound infection had their drain removed 1 week or longer after surgery (30.3% vs. 2.9%; P ¼ 0.002). We found no significant difference between cases and controls in terms of total drainage, average drainage, or drainage on the day of drain removal. Lastly, we found no difference in the percentage of drains removed with output greater than 100 mL or 200 mL in the prior 24 hours. On multivariable analysis of the 34 concordant pairs (Table 2), total drain indwelling time (odds ratio ¼ 1.36 per day; P ¼ 0.02) was a significant and independent predictor of postoperative wound infection requiring reoperation. Diabetes mellitus approached but did not reach statistical significance (P ¼ 0.16). Of note, lower preoperative hemoglobin level also predicted
WORLD NEUROSURGERY -: e1-e8, - 2019
DRAIN TIME AND SPINE SURGERY
development of postoperative wound infection (odds ratio ¼ 0.53 per g/dL increase in Hgb; P ¼ 0.005). None of the other variables found to be significant on univariable analysis proved significant on the multivariable analysis. A receiver operating curve constructed using these variables an associated C-statistic of 0.85 (Figure 2). Of infected patients (Table 3), the mean time between index surgery and readmission was 25.3 11.5 days (range: 8e53 days). A total of 33 patients (86.8%) had cultures grow out on their surgical specimens. The most commonly identified infecting agents were S aureus (33.3%), coagulase-negative Staphylococcus (15.2%), and polymicrobial gram negative cultures (12.1%). Of patients infected with S aureus, 4 grew out methicillin-resistant cultures (36.4%). DISCUSSION SSIs are debilitating postoperative complications that increase readmission rates,22,23 prolong hospital stays,3,4 increase overall morbidity,1,4 and produce worse patient outcomes.3,6 SSIs requiring surgical revision also increase direct care costs, with previous estimates varying between $15,000 and $38,000 depending upon the size of the index procedure.1,6,24 Because of this, there is great emphasis placed upon interventions designed to decrease surgical site infection, including placement of vancomycin powder in the surgical site,25 use of preoperative chlorhexidine wipes,26 and intravenous administration of antibiotics during the immediate pre-incisional period.27 Here, our results suggest that expediting surgical drain discharge may be another potential intervention as we found quicker surgical drain discharge was associated with a lower prevalence of deep SSI. A patient’s cumulative risk of surgical site infection is created by the combination of 2 categories of risk factors: 1) fixed factors, such as a patient’s medical morbidities (smoking, body mass index, diabetes mellitus type 2) and 2) modifiable factors, which include surgery duration, surgical technique, and perioperative and postoperative management. The latter category is that which is targeted by efforts to reduce infection risk and it includes the usage and timing of drain removal. There is a paucity of evidence regarding the degree to which surgical drain usage affects the incidence of spinal surgical site infection. Investigations in the orthopedic surgery literature have shown that prolonged surgical drains increase the risk of surgical site infection.28 This is speculated to stem from retrograde transmission of bacteria along the drain tract.29 Some small neurosurgical studies have not been able to replicate the link between drain usage and surgical site infection.12,13 Based on these studies, it is unclear whether drain indwelling time, output thresholds for drain removal, and average drain output affect the risk of SSI. Our data herein suggest that prolonged surgical drain indwelling time increases the risk for surgical site infection, independent of body habitus, surgical invasiveness, and medical history of diabetes mellitus. To our knowledge, only 2 prior studies have examined drain output and/or indwelling time in patients undergoing surgery for degenerative spine disease. Macki et al. examined a series of patients undergoing surgery for degenerative conditions of the
www.journals.elsevier.com/world-neurosurgery
e3
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
Table 1. Demographics and Comparison of Patients Requiring Surgical Revision for Surgical Site Infection Infection
No Infection
38
38
59.9 12.6
62.3 11.5
Male
28.9
39.5
Female
71.1
60.5
Variable n Age, years Sex
Race
Table 1. Continued Infection
No Infection
P Value
No drain
13.2
7.9
—
Variable P Value
Drain d/c POD1-2
10.5
18.4
—
0.02
Drain d/c POD3-4
31.6
52.6
—
0.33
Drain d/c POD 5e7
23.7
21.1
—
Drain d/c after POD 7
21.1
7.9
—
Drain d/c 7d
30.3
2.9
0.002
0.53
Drain d/c < 5d
51.5
77.1
0.02
1241.5 1102.9
912.5 672.2
0.11
Peak drainage, mL/24 hours
359.7 287.1
381.3 244.8
0.68
Average drainage, mL/24 hours
179.5 123.1
197.2 127.3
0.51
64.5 68.9
65.5 86.5
0.96
Drainage on last day 200 mL
24.2
17.1
0.34
Drainage on last day 100 mL
24.2
17.1
0.34
Drainage on last day 50 mL
42.4
37.1
0.42
9.5 10.9
4.3 2.4
0.005
42.4
45.7
0.49
White
71.1
78.9
Total drainage, mL
Black
21.1
18.4
Asian
2.6
2.6
Other
5.3
0
BMI, kg/m2
34.2 8.3
29.9 4.5
0.001
55.3
18.4
0.002
15.8
7.9
0.48
Medical history Diabetes Insulin-dependent Smoking status
0.45
Current
23.7
13.2
Former
23.7
31.6
52.6
55.3
Chronic kidney disease
Never
10.5
5.3
0.67
Congestive heart failure
5.3
5.3
1.0
COPD
15.8
5.3
0.26
HTN
81.6
57.9
0.04
Cancer*
7.9
5.3
1.0
Chronic steroid use
7.9
0
0.24
0
0
1.0
Coagulopathy
Length of stay, days
Surgery data Index surgery
0.03
Index
47.4
78.9
Revision
52.6
21.1
ASA score
0.03
1e2
13.2
36.8
3e4
86.8
63.2
19.3 14.1
18.3 13.6
0.14
278.1 112.9
280.6 128.8
0.86
86.8
92.1
0.71
5.49 4.83
3.51 1.34
0.02
ASD-S score Surgery duration, minutes Postoperative course Drain placed Drain retention time, days
Continues
e4
www.SCIENCEDIRECT.com
Drainage on last day, mL/24 hours
Postop abx given
Values are presented as mean standard deviation or %. Abx, antibiotics; ASA, American Society of Anesthesiologists; ASD-S, adult spinal deformity index; BMI, body mass index; BUN, blood urea nitrogen; COPD, chronic obstructive pulmonary disease; d/c, discharge; HTN, hypertension; POD, postoperative day. *Active malignancy in previous 5 years.
lumbar spine and found that neither drain removal setting (outpatient vs. inpatient) nor drain indwelling time were correlated with the rate of surgical revision for wound infection.30 However, among their cohort it was noted that those with drain retention time in the top quintile had an odds of surgical wound revision more than 6-fold higher than those with the median drain indwelling time, even after accounting for a history of diabetes mellitus type 2 and other comorbidities. Accordingly, it is possible that their lack of significant findings may be related to their small sample size. Herrick et al. examined a multiinstitutional series of patients undergoing posterior cervical spine surgery.31 They noted that although drain placement was an independent, negative predictor of return to the operating room, daily and total drain output were greater in the group of patients requiring reoperation. Their data further showed that patients requiring revision operation had higher drain indwelling time (3.58 vs. 2.34 days). Similar to the study by Macki et al., the authors found no significant relationship between drain indwelling time and the need for reoperation for wound
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
Table 2. Multivariable Logistic Compression Predicting Patients Requiring Surgical Revision for Wound Infection Among Those Being Treated for Degenerative Spine Pathologies with Postoperative Drain Placement 95% CI Predictor
OR
Lower
Upper
P Value
BMI (per kg/m2)
1.18
1.06
1.33
0.004
Hgb (per g/dL)
0.53
0.35
0.82
0.005
Drain retention time (per day)
1.36
1.04
1.78
0.02
ASA, American Society of Anesthesiologists; BMI, body mass index; CI, confidence interval; Hgb, hemoglobin; OR, odds ratio.
infection. Given that infected patients had drains maintained for an average of 1.2 days longer than non-infected patients, it can be suggested that there may have been a trend that just did not reach statistical significance. Given this limited study of the linkage between drain indwelling time and infection risk, the manner in which prolonged drain indwelling increases the risk of surgical site infection is unknown. However, evidence from both animal experiments32,33 and clinical orthopedic literature has suggested that placement of a surgical
Figure 2. Receiver-operating curve demonstrating the accuracy of the multivariable logistic regression
WORLD NEUROSURGERY -: e1-e8, - 2019
drain may enable the tracking of skin flora and other epidermal bacteria into the surgical wound.34 Using rabbit models, both Raves et al.32 and Cerise et al.33 found that placement of closed suction drains (Jackson-Pratt drains) led to significantly higher rates of deep wound infection. Additionally, Cerise et al. found drain placement to be associated with a significant increase in deep drain site abscess formation among 533 patients undergoing splenectomy. Sørensen and Sørensen documented that among 500 orthopedic patients, positive drain tip cultures was associated with an increased risk of infection, as was a drain indwelling time greater than 6 days.34 Similarly, research from the field of reconstructive surgery has found bacterial colonization of surgical drains to increase substantially with prolonged drain indwelling time, with 1 study finding that drain colonization rates tripled from postoperative day 7 to day 14.35 More recently, similar results have been reported in breast surgery, revealing that each additional week of drain indwelling time increases the risk of surgical site infection by 76.2%.36 In aggregate these studies suggest that early drain removal should decrease wound site penetration by the skin microfauna, lowering the risk of deep SSI and commensurately lowering the likelihood of requiring surgical wound revision.37 Despite these potential negative sequelae of surgical drains, other prior studies have found drain use to be associated with a significantly lower rate of wound infection among both degenerative and non-degenerative spine patients.31,38 Additionally, many
incorporating drain indwelling time as an independent predictor of surgical site infection requiring reoperation.
www.journals.elsevier.com/world-neurosurgery
e5
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
Table 3. Results of Sample Cultures for Infected Patients Case
Time to Infection
Culture Result
Gram Stain
Infecting Agent
Length of Stay (days)
1
22
Pos
Pos
MRSA
59
2
31
Neg
N/A
Unknown
6
3
17
Pos
Neg
Polymicrobial gram negative
25
4
10
Pos
Pos
Coagulase-negative Staphylococcus
3
5
24
Neg
N/A
Unknown
23
6
42
Pos
Neg
B. thetaiotaomicron
32
7
20
Pos
Neg
E. coli
15
8
22
Pos
Pos
E. faecalis
5
9
15
Pos
Pos
P. acnes
7
10
19
Pos
Pos
MSSA
23
11
26
Pos
Neg
Polymicrobial gram negative
10
12
33
Pos
Pos
MSSA
7
13
46
Pos
Pos
E. faecalis
13
14
36
Pos
Pos
MRSA
7
15
15
Pos
Pos
MRSA
9
16
20
Pos
Neg
E. coli
5
17
21
Pos
Pos
E. faecalis, S. marcescens
19
18
29
Pos
Neg
P. mirabilis
6
19
15
Neg
N/A
Unknown
36
20
21
Pos
Pos
Coagulase-negative Staphylococcus
29
21
34
Pos
Neg
P. mirabilis
7
22
36
Pos
Pos
MSSA
26
23
30
Pos
Pos
Coagulase-negative Staphylococcus
6
24
14
Pos
Pos
MSSA
6
25
31
Pos
Pos
Coagulase-negative Staphylococcus
9
26
39
Pos
Pos
MSSA
7
27
33
Pos
Pos
MSSA
5
28
13
Pos
Pos
S. pneumoniae
31
29
17
Neg
N/A
Unknown
4
30
53
Pos
Pos
MRSA
13
31
15
Pos
Pos
MSSA
2
32
19
Pos
Neg
Polymicrobial gram negative
17
33
14
Pos
Neg
E. coli
11
34
11
Pos
Neg
Polymicrobial gram negative
9
35
47
Pos
Pos
Group B Streptococcus
3
36
20
Pos
Neg
S. marcescens
7
37
43
Pos
Pos
Coagulase-negative Staphylococcus
7
38
8
Neg
N/A
Unknown
11
MRSA, methicillin-resistant staphylococcus aureus; MSSA, methicillin-sensitive staphylococcus aureus; N/A, not available; Neg, negative; Pos, positive.
e6
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
DRAIN TIME AND SPINE SURGERY
surgeons elect to place a drain for reasons other than infection prophylaxis, including the potential to decrease the risk of wound site hematoma and subsequent neurological compromise.31 The bulk of the evidence has failed to support the utility of drains for this purpose, however,39 although many groups still advocate for removal of the drain only after output drops below 50 mL per day.40-42 Given the lack of evidence in the literature and our observation of no correlation between SSI and final day drain output up to 200 mL per 24-hour period, we believe that for the average patient undergoing surgery for degenerative spine disease, drains may safely be removed while still outputting as much as 200 mL per day. In addition to potentially reducing the risk of SSI, early drain discharge would likely decrease total care costs. The time required for drain output to drop from 100 mL/24 hours to 50 mL/24 hours may prolong hospital stays by an additional 1 or 2 days, which according to the Kaiser Foundation translates into a cost increase of $1,889 to $4,976.43 Assuming the lower estimate and extrapolating to the entire US spinal surgery population (estimated at 500,000 people annually44) we find that total savings to the US health system could approach $500 million to $1 billion. Limitations There are several limitations to this study, including the small sample size. Our study examined only patients treated for the primary indication of SSI after surgery for degenerative conditions of the spine. Small sample populations are more susceptible to bias by outlier observations. In the present cohort one such bias may stem from the inconsistent application of vancomycin powder to the operative wound across surgeons. The proportion of patients receiving this intervention, along with other occult confounders, potentially limit the generalizability of the conclusions that can be drawn from them. Because of this, it
2. Smith JS, Shaffrey CI, Sansur CA, et al. Rates of infection after spine surgery based on 108,419 procedures: a report from the Scoliosis Research Society Morbidity and Mortality Committee. Spine (Phila Pa 1976). 2011;36:556-563. 3. Badia JM, Casey AL, Petrosillo N, Hudson PM, Mitchell SA, Crosby C. Impact of surgical site infection on healthcare costs and patient outcomes: a systematic review in six European countries. J Hosp Infect. 2017;96:1-15. 4. Casper DS, Zmistowski B, Hollern DA, et al. The effect of postoperative spinal infections on patient mortality. Spine (Phila Pa 1976). 2018;43:223-227. 5. Berríos-Torres SI, Umscheid CA, Bratzler DW, et al. Centers for Disease Control and Prevention Guideline for the prevention of surgical site infection, 2017. JAMA Surg. 2017;152:784. 6. Kuhns BD, Lubelski D, Alvin MD, et al. Cost and quality of life outcome analysis of postoperative
CONCLUSIONS Longer drain indwelling time seems to be an independent risk factor for SSIs after surgery for degenerative spine pathologies. Most importantly, drain retention for greater than 1 week appears to be associated with a higher rate of SSIs. To our knowledge, this is the first study to directly correlate drain indwelling time with the occurrence of spinal wound site infections. The results of the present study suggest that drain discontinuation may be considered earlier to reduce infections and decrease length of hospital stay. Using this higher threshold may decrease drain indwelling times and decrease the rate of SSIs among spine surgery patients. Additionally, simple cost analysis suggests that expediting drain discharge may decrease overall care costs within this population.
infections after subaxial dorsal cervical fusions. J Neurosurg Spine. 2015;22:381-386.
REFERENCES 1. Anderson PA, Savage JW, Vaccaro AR, et al. Prevention of surgical site infection in spine surgery. Neurosurgery. 2017;80(3 Suppl):S114-S123.
is possible that our results may not be applicable to all surgical patients. Patients had skin preparation with either chlorhexidinebased or povidone-iodineebased agents as well as iodophoreimpregnated drapes. The evidence is equivocal regarding the superiority of one preparation type versus another and this may potentially lead to variability in incision site colonization and possible infection risk. The small sample size may also reduce the sensitivity of our analysis by hampering our ability to identify true risk factors that have relatively small contributions to a patient’s cumulative risk. Additionally, our study is retrospective in nature, meaning that we are only able to correlate drain indwelling time with the odds of having experienced an SSI; we are unable to determine whether this relationship is causal. In future studies, we hope to use a larger patient cohort from multiple institutions to provide more robust results. Lastly, given the high proportion of patients in both cohorts who received a surgical drain, our study may have had insufficient power to identify the contribution of drain placement versus nonplacement to the occurrence of deep SSI.
7. Meng F, Cao J, Meng X. Risk factors for surgical site infections following spinal surgery. J Clin Neurosci. 2015;22:1862-1866. 8. Yao R, Zhou H, Choma TJ, Kwon BK, Street J. Surgical site infection in spine surgery: who is at risk? Global Spine J. 2018;8(4 suppl):5S-30S. 9. Mehta AI, Babu R, Sharma R, et al. Thickness of subcutaneous fat as a risk factor for infection in cervical spine fusion surgery. J Bone Joint Surg Am. 2013;95:323-328. 10. Mehta AI, Babu R, Karikari IO, et al. 2012 Young Investigator Award winner: the distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine (Phila Pa 1976). 2012;37:1652-1656. 11. Chimenti P, Molinari R. Post-operative spinal epidural hematoma causing American Spinal Injury Association B spinal cord injury in patients with suction wound drains. J Spinal Cord Med. 2013; 36:213-219.
WORLD NEUROSURGERY -: e1-e8, - 2019
12. Andrew Glennie R, Dea N, Street JT. Dressings and drains in posterior spine surgery and their effect on wound complications. J Clin Neurosci. 2015;22:1081-1087. 13. Liu Y, Li Y, Miao J. Wound drains in posterior spinal surgery: a meta-analysis. J Orthop Surg Res. 2016;11:16. 14. Chen Z, Gao Y, Chen W, Li X, Zhang Y. Is wound drainage necessary in hip arthroplasty? A metaanalysis of randomized controlled trials. Eur J Orthop Surg Traumatol. 2014;24:939-946. 15. Zhang Q, Guo W, Zhang Q, Liu Z, Cheng L, Li Z. Comparison between closed suction drainage and nondrainage in total knee arthroplasty: a metaanalysis. J Arthroplasty. 2011;26:1265-1272. 16. Adogwa O, Elsamadicy AA, Sergesketter AR, et al. Post-operative drain use in patients undergoing decompression and fusion: incidence of complications and symptomatic hematoma. J Spine Surg. 2018;4:220-226. 17. Adogwa O, Khalid SI, Elsamadicy AA, et al. The use of subfascial drains after multi-level anterior cervical discectomy and fusion: does the data support its use? J Spine Surg. 2018;4:227-232.
www.journals.elsevier.com/world-neurosurgery
e7
ORIGINAL ARTICLE ZACH PENNINGTON ET AL.
18. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Am J Infect Control. 1999;27: 97-134. 19. Neuman BJ, Ailon T, Scheer JK, et al. Development and validation of a novel adult spinal deformity surgical invasiveness score: analysis of 464 patients. Neurosurgery. 2018;82:847-853. 20. Pawar AY, Biswas SK. Postoperative spine infections. Asian Spine J. 2016;10:176-183. 21. Ranson WA, White SJW, Cheung ZB, et al. The effects of chronic preoperative steroid therapy on perioperative complications following elective posterior lumbar fusion. Glob Spine J. 2018;8: 834-841. 22. Lee JJ, An SB, Kim TW, et al. Analysis of risk factors associated with hospital readmission within 360 days after degenerative lumbar spine surgery in elderly patients. World Neurosurg. 2019: e196-e207. 23. Park P, Nerenz DR, Aleem IS, et al. Risk factors associated with 90-day readmissions after degenerative lumbar fusion: an examination of the Michigan Spine Surgery Improvement Collaborative (MSSIC) Registry. Neurosurgery. 2018:1-7. 24. McGirt MJ, Parker SL, Lerner J, Engelhart L, Knight T, Wang MY. Comparative analysis of perioperative surgical site infection after minimally invasive versus open posterior/transforaminal lumbar interbody fusion: analysis of hospital billing and discharge data from 5170 patients. J Neurosurg Spine. 2011;14:771-778. 25. He X, Sun T, Wang J, Li G, Fei Q. Application of vancomycin powder to reduce surgical infection and deep surgical infection in spinal surgery: a meta-analysis. Clin Spine Surg. 2019;32:150-163. 26. Savage JW, Anderson PA. An update on modifiable factors to reduce the risk of surgical site infections. Spine J. 2013;13:1017-1029. 27. Barker FG II. Efficacy of prophylactic antibiotic therapy in spinal surgery: a meta-analysis. Neurosurgery. 2002;51:391-401. 28. Saleh K, Olson M, Resig S, et al. Predictors of wound infection in hip and knee joint
e8
www.SCIENCEDIRECT.com
DRAIN TIME AND SPINE SURGERY
replacement: results from a 20 year surveillance program. J Orthop Res. 2002;20:506-515. 29. Reiffel AJ, Barie PS, Spector JA. A multi-disciplinary review of the potential association between closed-suction drains and surgical site infection. Surg Infect (Larchmt). 2013;14:244-269.
39. Wang L, Liu X. Cervical sagittal alignment in adolescent idiopathic scoliosis patients (Lenke type 1-6). J Orthop Sci. 2017;22:254-259. 40. Kanayama M, Oha F, Togawa D, Shigenobu K, Hashimoto T. Is closed-suction drainage necessary for single-level lumbar decompression?: review of 560 cases. Clin Orthop Relat Res. 2010;468: 2690-2694.
30. Macki M, Uzosike A, Kerezoudis P, Bydon A, Bydon M, Gokaslan ZL. Duration of indwelling drain following instrumented posterolateral fusion of the lumbar spine does not predict surgical site infection requiring reoperation. J Clin Neurosci. 2017;40:44-48.
41. Diab M, Smucny M, Dormans JP, et al. Use and outcomes of wound drain in spinal fusion for adolescent idiopathic scoliosis. Spine (Phila Pa 1976). 2012;37:966-973.
31. 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;29:1-7.
42. Kamath VHD, Cheung JPY, Mak KC, et al. Antimicrobial prophylaxis to prevent surgical site infection in adolescent idiopathic scoliosis patients undergoing posterior spinal fusion: 2 doses versus antibiotics till drain removal. Eur Spine J. 2016;25:3242-3248.
32. Raves JJ, Slifkin M, Diamond DL. A bacteriologic study comparing closed suction and simple conduit drainage. Am J Surg. 1984;148:618-620.
43. Henry J. Kaiser Family Foundation. Hospital adjusted expenses per inpatient day by ownership. Available at: https://www.kff.org/health-costs/ state-indicator/expenses-per-inpatient-day-by-own ership/?currentTimeframe¼0&sortModel¼%7B% 22colId%22:%22Location%22,%22sort%22:%22asc% 22%7D; 2016. Accessed February 17, 2019.
33. Cerise EJ, Pierce WA, Diamond DL. Abdominal drains: their role as a source of infection following splenectomy. Ann Surg. 1970;171:764-769. 34. Sørensen AI, Sørensen TS. Bacterial growth on suction drain tips. Prospective study of 489 clean orthopedic operations. Acta Orthop Scand. 1991;62: 451-454. 35. Felippe WAB, Werneck GL, Santoro-Lopes G. Surgical site infection among women discharged with a drain in situ after breast cancer surgery. World J Surg. 2007;31:2293-2299. 36. Chen C-F, Lin S-F, Hung C-F, Chou P. Risk of infection is associated more with drain duration than daily drainage volume in prosthesis-based breast reconstruction: a cohort study. Medicine (Baltimore). 2016;95:e5605. 37. Kobayashi K, Imagama S, Ito Z, et al. Is a drain tip culture required after spinal surgery? Clin Spine Surg. 2017;30:356-359. 38. Ramo BA, Roberts DW, Tuason D, et al. Surgical site infections after posterior spinal fusion for neuromuscular scoliosis. J Bone Joint Surg Am. 2014; 96:2038-2048.
44. Steiner CA, Karaca Z, Moore BJ, Imshaug MC, Pickens G. Surgeries in Hospital-Based Ambulatory Surgery and Hospital Inpatient Settings, 2014. Rockville, MD: Agency for Healthcare Research and Quality; 2014. Conflict of interest statement: C. Molina is a consultant for Augmedics. D.M. Sciubba is a consultant for Baxter, DePuySynthes, Globus, K2M, Medtronic, NuVasive, and Stryker. The remaining authors have no conflicts to report. Received 29 April 2019; accepted 1 July 2019 Citation: World Neurosurg. (2019). https://doi.org/10.1016/j.wneu.2019.07.013 Journal homepage: www.journals.elsevier.com/worldneurosurgery Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2019 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.07.013