1732

C OPYRIGHT Ó 2019 BY T HE J OURNAL OF B ONE AND J OINT S URGERY, I NCORPORATED

The Effect of Prolonged Postoperative Antibiotic

Administration on the Rate of Infection in Patients
Undergoing Posterior Spinal Surgery Requiring a
Closed-Suction Drain
A Randomized Controlled Trial
Jennifer C. Urquhart, PhD, Darryl Collings, MD, Lori Nutt, RN, Linda Kuska, RN, Kevin R. Gurr, MD, FRCSC,
Fawaz Siddiqi, MD, FRCSC, Parham Rasoulinejad, MD, FRCSC, Alyssa Fleming, Joanne Collie,
and Christopher S. Bailey, MD, MSc, FRCSC

Investigation performed at London Health Sciences Centre, London, Ontario, Canada

Background: Closed-suction drains are frequently used following posterior spinal surgery. The optimal timing of anti-
biotic discontinuation in this population may influence infection risk, but there is a paucity of evidence. The aim of this
study was to determine whether postoperative antibiotic administration for 72 hours (24 hours after drain removal as
drains were removed on the second postoperative day) decreases the incidence of surgical site infection compared with
postoperative antibiotic administration for 24 hours.
Methods: Patients undergoing posterior thoracolumbar spinal surgery managed with a closed-suction drain were pro-
spectively randomized into 1 of 2 groups of postoperative antibiotic durations: (1) 24 hours, or (2) 24 hours after drain
removal (72 hours). Drains were discontinued on the second postoperative day. The duration of antibiotic administration
was not blinded. All subjects received a single dose of preoperative antibiotics, as well as intraoperative antibiotics if the
surgical procedure lasted >4 hours. The primary outcome was the rate of complicated surgical site infection (deep or organ
or space) within 1 year of the surgical procedure.
Results: The trial was terminated at an interim analysis, when 552 patients were enrolled, for futility with respect to the
primary outcome. In this study, 282 patients were randomized to postoperative antibiotics for 24 hours and 270 patients
were randomized to postoperative antibiotics for 72 hours. A complicated infection developed in 17 patients (6.0%) in the
24-hour group and in 14 patients (5.2%) in the 72-hour group (p = 0.714). The superficial infection rate did not differ
between the groups (p = 0.654): 9.6% in the 24-hour group compared with 8.1% in the 72-hour group. Patients in the 72-
hour group had a median hospital stay that was 1 day longer (p < 0.001). At 1 year, patient-rated outcomes including leg
and back pain and physical and mental functioning were not different between the groups.
Conclusions: The extension of postoperative antibiotics for 72 hours, when a closed-suction drain is required, was not
associated with a reduction in the rate of complicated surgical site infection after posterior thoracolumbar spinal surgery.
Level of Evidence: Therapeutic Level I. See Instructions for Authors for a complete description of Levels of Evidence.

T
he rate of surgical site infection after a posterior tho- admission, an irrigation and debridement surgical procedure,
racolumbar instrumented surgical procedure is and long-term intravenous antibiotics5. Furthermore, the risk
reportedly 9% to 18%1-4. Treatment often requires re- of pseudarthrosis and instrument-related complications is

Disclosure: The authors indicated that no external funding was received for any aspect of this work. On the Disclosure of Potential Conflicts of Interest
forms, which are provided with the online version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial
relationship in the biomedical arena outside the submitted work (http://links.lww.com/JBJS/F419).

A data-sharing statement is provided with the online version of the article (http://links.lww.com/JBJS/F420).

J Bone Joint Surg Am. 2019;101:1732-40 d http://dx.doi.org/10.2106/JBJS.19.00009

 1733
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
increased, leading to considerable personal and financial costs
to the patient, as well as increased resource utilization and
economic cost to the health-care system5,6.
Closed-suction drains are frequently used after spinal
surgery to circumvent the risk of neurological deterioration
from a compressive hematoma7-9. Additionally, hematoma pre-
vention may reduce the risk of infection by eliminating the
medium for bacterial growth8. In contrast to these proposed
benefits, drains have been associated with an increase in the
surgical site infection rate10-13.
Clinical practice guidelines have recommended antibi-
otic prophylaxis to reduce surgical site infection14-20. For spinal
procedures with instrumentation, it is recommended that
antibiotics be discontinued within 24 hours of the surgical end
time20,21. The North American Spine Society guidelines stated
that there is insufficient evidence to make a recommendation
regarding the duration of prophylaxis in the presence of a wound
drain and suggested that controlled studies are needed16. To date,
only 1 Level-I study has shown that extending postoperative
antibiotic coverage for the duration that a drain is in place does
not affect the surgical site infection rate after posterior multilevel
thoracolumbar spinal surgery22. The study did not access
whether prolonged prophylaxis after drain removal decreased
infection risk.
An open, posterior thoracolumbar surgical procedure,
which requires instrumentation or a multilevel surgical pro-
cedure, is at a higher risk of postoperative wound infections.
Specifically for this population, further high-quality evidence is
required to support decisions about antibiotic discontinuation
for procedures in which a closed-suction drain is utilized,
particularly when considering that prolonged antibiotic use
Fig. 1
Flow diagram of recruitment and randomization. h = hours and pre-op = preoperatively.
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N
increases the risk of nosocomial infections, bacterial resistance,
and cost23,24.
The purpose of the present study was to compare the rate
of complicated surgical site infection after posterior, open
thoracolumbar spine procedures followed by the placement of
a closed-suction drain between patients treated with postop-
erative antibiotics for 24 hours and 72 hours. The rationale for
choosing postoperative antibiotics for 72 hours was to ensure
that patients received 24 hours of postoperative antibiotics after
drain removal, as drains were discontinued by postoperative
day 2. We hypothesized that the complicated infection rate
would be higher in the patients who received the shorter
duration of postoperative antibiotic prophylaxis.
Materials and Methods
Study Setting and Population
A consecutive series of patients were assessed for eligibility at
a single center between October 2011 and April 2016.
Inclusion criteria were age of ‡16 years; elective, open posterior
thoracic or lumbar multilevel decompression and/or arthrodesis
for deformity or degenerative conditions, of large enough
magnitude that the insertion of a closed-suction drain would be
considered; and ability to provide informed consent (English
speaking). Exclusion criteria were known antibiotic hypersen-
sitivity; creatinine level of >100 mmol/L; antibiotic therapy for
concomitant infection; a surgical procedure for primary spinal
infection, spinal tumor, or trauma; pregnancy; concomitant
corticosteroid therapy; and a permanent residence with a driving
distance to a hospital of >5 hours. This study was approved by
our institutional research ethics board. The protocol was regis-
tered with ClinicalTrials.gov (NCT01458223).
 1734
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N

Study Design
Participating surgeons assessed eligibility. The study coordi-
nator then enrolled and obtained consent from all patients.
Using a computer randomization program, patients were
randomly assigned by 1:1 allocation to antibiotic administra-
tion for 24 hours after the operation or for 24 hours after the
drain removal (72 hours after the operation). The stratifying
factors were the surgeon and the presence or absence of dia-
betes, to ensure equal balance between treatment groups. The
allocation was concealed from the surgeon.
One of 4 fellowship-trained spine surgeons performed the
operations. The operative technique was chosen at the surgeon’s
discretion. All procedures were open, posterior approaches uti-
lizing a multilevel laminectomy and/or pedicle screw and rod
instrumentation, with or without an interbody device.
Patients received 2 g of cefazolin intravenously, 1 hour to
no later than 15 minutes prior to the surgical incision, or 1 g of
vancomycin within 1 hour of incision if the patient was allergic to
penicillin or cephalosporin. A second dose of cefazolin was given
at the 4-hour mark to patients without a cefazolin allergy if the
surgical procedure lasted >4 hours21. Prior to closure, patients
had wound irrigation with 1 L of warmed normal saline solution,
and a 10-French (0.13-inch) Jackson Pratt Wound Drain (Car-
dinal Health) was inserted into the subfascial space. If a vacuum

TABLE I Patient Baseline Demographic Characteristics

Overall Cohort 24-Hour Group 72-Hour Group

Parameter (N = 552) (N = 282) (N = 270) P Value

Age* (yr) 58.9 ± 14.5 57.6 ± 14.8 60.2 ± 14.2 0.060

Female sex† 296 (53.6) 151 (53.5) 145 (53.7) 1.00
Body mass index* (kg/m2) 29.1 ± 5.9 29.1 ± 5.8 29.0 ± 5.9 0.979
Diabetes† 91 (16.5) 48 (17.0) 43 (15.9) 0.733
Smoker† 0.407
No 314 (56.9) 156 (55.3) 158 (58.5)
Yes 128 (23.2) 65 (23.0) 63 (23.3)
No longer 71 (12.9) 36 (12.8) 35 (13.0)
Chose not to answer 39 (7.1) 25 (8.9) 14 (5.2)
Primary diagnoses† 0.535
Deformity 12 (2.2) 6 (2.1) 6 (2.2)
Disc herniation‡ 112 (20.3) 66 (23.4) 46 (17.0)
Spinal stenosis 79 (14.3) 46 (16.3) 33 (12.2)
Spinal stenosis with scoliosis or kyphosis 18 (3.3) 10 (3.5) 8 (3.0)
Degenerative spondylolisthesis 228 (41.3) 98 (34.8) 130 (48.1)
Isthmic spondylolisthesis 103 (18.7) 56 (19.9) 47 (17.4)
Comorbidities† §
None 106 (19.2) 58 (20.6) 48 (17.8) 0.450
Anemia or other blood disease 35 (6.3) 14 (5.0) 21 (7.8) 0.333
Hypertension 79 (14.3) 34 (12.1) 45 (16.7) 0.269
Cerebrovascular disease 15 (2.7) 7 (2.5) 8 (3.0) 0.726
Joint disease (osteoarthritis) 108 (19.6) 62 (22.0) 46 (17.0) 0.193
Heart disease 18 (3.3) 13 (4.6) 5 (1.9) 0.128
Lung disease 47 (8.5) 25 (8.9) 22 (8.1) 0.700
Gastrointestinal disease 15 (2.7) 6 (2.1) 9 (3.3) 0.541
Liver disease (mild or moderate) 7 (1.3) 2 (0.7) 5 (1.9) 0.384
History of cancer 29 (5.3) 12 (4.3) 17 (6.3) 0.418
Depression 98 (17.8) 50 (17.7) 48 (17.8) 0.751
Other 10 (1.8) 5 (1.8) 5 (1.9) 0.758
Inflammatory arthritis 3 (0.5) 2 (0.7) 1 (0.4) 0.646
Unknown or choose not to answer 81 (14.7) 45 (16.0) 36 (13.3) 0.402

*The values are given as the mean and the standard deviation. †The values are given as the number of patients, with the percentage in
parentheses. ‡This includes recurrent, intraforaminal, and multilevel disc herniation, as well as central protrusions producing spinal
stenosis. §Some patients had >1 comorbid condition.
 1735
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N

wound drain was contraindicated, the patient was excluded.
Dressings were first changed after 48 hours and thereafter on an
as-needed basis using sterile technique. Postoperatively, patients
received either 1 g of cefazolin every 8 hours or 1 g of vancomycin
every 12 hours according to their randomization. In both groups,
drain removal was standardized to approximately 48 hours
postoperatively. Physicians did not administer the antibiotics and
were not blinded to the assigned treatment.
The primary outcome measure was a complicated sur-
gical site infection, defined as a deep or organ or space infec-
tion, within 1 year. The U.S. Centers for Disease Control and
Prevention definitions for superficial, deep, and organ or space
surgical site infection were used in the diagnosis25. Deep in-
fections were culture-positive and/or contained purulent ma-
terial. Demographic characteristics, diagnosis, procedure, and
adverse events were collected during the initial hospitalization.
Any surgical site infection or other infections were documented
prior to discharge. At 2 weeks, patients were contacted by
telephone to screen for wound drainage or prescribed antibi-
otics. If there was a concern, their surgeon inspected the sur-
gical site. Otherwise, patients underwent follow-up at 6 weeks,
3 months, 6 months, and 12 months.
Validated patient-rated outcome measures were admin-
istered preoperatively at the regularly scheduled visits stated
above. Leg and back pain intensity was assessed with a nu-
merical rating scale, 0 = no pain to 10 = worst imaginable
pain26. The Oswestry Disability Index (ODI) was used to mea-
sure disability27. The Medical Outcomes Study 12-item Short-
Form General Health Survey (SF-12) Physical Component
Summary (PCS) and Mental Component Summary (MCS)
scores were used to evaluate health-related quality of life28.
Adverse events were collected using Spine Adverse Events
Severity System version 2 (SAVES-V2)29.
Statistical Analysis
The overall mean deep surgical site infection rate at our in-
stitution from 2008 to 2010 was 5.5%. We assumed that the
intervention would reduce the infection rate by 3%18. Assuming
a power of 0.80 and an alpha of 0.05, a minimum of 533 per
group was needed using the chi-square test for 2 independent
proportions.
We elected to perform an interim analysis when half of
the planned number of subjects had been enrolled. The analysis
was blinded and was performed by an independent data safety
monitoring committee. To determine futility, the stochastic
curtailment method based on conditional power was used with
a prespecified level of conditional power of £0.230-32.
Data analysis was performed using PASW Statistics version
24 (SPSS). For the primary outcome, the rate of complicated
infection was compared using the chi-square test performed on
an intention-to-treat basis. For baseline characteristics and sec-
ondary outcome comparisons, between-group comparisons
were made using the Student t test for continuous parametric
variables or the Mann-Whitney U test for continuous nonpara-
metric variables. Comparisons for categorical variables were
made using the chi-square test or the Fisher exact test. Missing
patient-rated outcome measures were assumed to be missing at
random and were replaced with a set of plausible values derived
using a multiple imputation procedure with 10 iterations taking
all time points into consideration. An as-treated sensitivity
analysis was performed according to the duration of antibiotics
received. A significant difference was defined as p < 0.05.
Results
Patient Enrollment and Characteristics of the Patients
F igure 1 shows patient recruitment, randomization, and the
number of patients analyzed in the primary analysis and the

TABLE II Infection Rates, Time to Infection, and Other Wound Complications

Overall Cohort 24-Hour Group 72-Hour Group

Parameter (N = 552) (N = 282) (N = 270) P Value

Complicated infection* 31 (5.6) 17 (6.0) 14 (5.2)† 0.714

Superficial infection* 49 (8.9) 27 (9.6) 22 (8.1) 0.654
Other infections requiring antibiotic treatment 19 (3.4) 8 (2.8) 11 (4.1) 0.488
(i.e., urinary tract infection, pneumonia)*
Wound draining at 2 to 6 weeks with no 57 (10.3) 24 (8.5) 33 (12.2) 0.164
infection or antibiotic treatment*
Time to infection (days) 0.628
Mean and standard deviation 22.6 ± 38.1 18.9 ± 25.4 27.2 ± 49.5
Median (range) 15 (3 to 303) 15 (3 to 178) 16 (4 to 303)
Other wound complications*
Hematoma 11 (2.0) 8 (2.8) 3 (1.1) 0.223
Dehiscence 18 (3.3) 9 (3.2) 9 (3.3) 1.000

*The values are given as the number of patients, with the percentage in parentheses. †This includes 1 organ or space infection: adjacent-level
discitis treated at 303 days after the surgical procedure.
 1736
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N

TABLE III Type of Surgical Procedure, Perioperative Details, and Postoperative Outcomes

Overall Cohort 24-Hour Group 72-Hour

Parameter (N = 552) (N = 282) Group (N = 270) P Value

Antibiotic* 0.865
Cefazolin 515 (93.3) 264 (93.6) 251 (93.0)
Vancomycin 37 (6.7) 18 (6.4) 19 (7.0)
Type of surgical procedure* 0.105
Multilevel decompression 5 (0.9) 5 (1.8) 0 (0.0)
Instrumented arthrodesis 5 (0.9) 3 (1.1) 2 (0.7)
Decompression and arthrodesis
Interbody arthrodesis 458 (83.0) 236 (83.7) 222 (82.2)
Instrumented arthrodesis 84 (15.2) 38 (13.5) 46 (17.0)
No. of surgical levels† 2.6 (2 to 14) 2.5 (2 to 12) 2.7 (2 to 14) 0.111
Revision* 103 (18.7) 52 (18.4) 51 (18.9) 0.913
American Society of Anesthesiologists classification* 0.912
1 9 (1.6) 4 (1.4) 5 (1.9)
2 189 (34.2) 96 (34.0) 93 (34.4)
3 334 (60.5) 173 (61.3) 161 (59.6)
4 20 (3.6) 9 (3.2) 11 (4.1)
Duration of surgical procedure (min) 0.052
Mean and standard deviation 210 ± 60.2 204 ± 59.4 215 ± 60.7
Median (range) 205 (58 to 630) 203 (58 to 410) 209 (71 to 630)
Blood transfusion* 45 (8.2) 18 (6.4) 27 (10.0) 0.161
Estimated blood loss‡ (mL) 0.053
Mean and standard deviation 685.1 ± 408.9 652.9 ± 408.9 718.0 ± 441.9
Median (range) 600 (50 to 3,500) 600 (50 to 3,000) 600 (100 to 3,500)
Hospital stay (days) 0.0001
Mean and standard deviation 4.2 ± 2.9 3.9 ± 2.4 4.5 ± 3.3
Median (range) 4 (1 to 43) 3 (1 to 25) 4 (1 to 43)
Intensive care unit admission* 5 (0.9) 1 (0.4) 4 (1.5) 0.207
Adverse events* §
Implant or instrument-related 6 (1.1) 4 (1.4) 2 (0.7) 0.686
Instrument, fixation, implant, malposition 4 (0.7) 0 (0.0) 4 (1.5) 0.057
requiring revision
Myocardial infarction 2 (0.4) 1 (0.4) 1 (0.4) 1.000
Neurologic deterioration 5 (0.9) 3 (1.1) 2 (0.7) 1.000
Nonunion 1 (0.2) 0 (0.0) 1 (0.4) 0.489
New-onset pain 25 (4.5) 12 (4.3) 13 (4.8) 0.839
Urinary retention 3 (0.5) 3 (1.1) 0 (0.0) 0.249
Wound drainage, cerebrospinal fluid 8 (1.4) 4 (1.4) 4 (1.5) 1.000
Dural tear 49 (8.9) 24 (8.5) 25 (9.3) 0.767
Other 39 (7.1) 20 (7.1) 19 (7.0) 1.000
Additional surgical procedure including irrigation 57 (10.3) 35 (12.4) 22 (8.1) 0.123
and debridement*

*The values are given as the number of patients, with the percentage in parentheses. †The values are given as the mean, with the range in
parentheses. ‡Data were available for 263 patients in the 24-hour group and 256 patients in the 72-hour group. §Some patients had >1
adverse event.
 1737
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N

TABLE IV Baseline Patient-Rated Health Outcome Scores and the Change in Score from Baseline to 1 Year Postoperatively

Baseline (Preoperative) Score Change in Score*

Outcomes 24-Hour Group† 72-Hour Group† Difference‡ P Value 24-Hour Group† 72-Hour Group† Difference‡ P Value

ODI 49.5 ± 1.0 47.2 ± 1.1 22.3 (25.3 to 0.6) 0.126 218.8 ± 1.3 216.4 ± 1.2 2.4 (21.1 to 5.9) 0.175
Leg pain 7.6 ± 0.1 7.4 ± 0.2 20.2 (20.6 to 0.2) 0.347 24.3 ± 0.3 24.5 ± 0.3 0.04 (20.7 to 0.8) 0.908
Back pain 7.3 ± 0.1 7.2 ± 0.1 20.1 (20.5 to 0.3) 0.546 23.6 ± 0.2 23.6 ± 0.2 0.05 (20.6 to 0.7) 0.878
SF-12 PCS 28.3 ± 0.5 28.3 ± 0.5 20.01 (21.3 to 1.3) 0.991 9.2 ± 0.7 9.0 ± 0.6 20.2 (22.1 to 1.7) 0.845
SF-12 MCS 42.9 ± 0.7 44.1 ± 0.8 1.2 (20.9 to 3.3) 0.253 6.7 ± 0.9 6.4 ± 0.8 20.3 (22.6 to 2.0) 0.809

*Data were derived using multiple imputation to impute missing data. †The values are given as the mean and the standard error in points. ‡The values are given as the mean and the
95% confidence interval in points. For the PCS and MCS, a positive change indicated improvement. For the ODI score, back pain score, and leg pain score, a negative change indicated
improvement.

as-treated analysis. Three patients in the 24-hour group
received >24 hours of postoperative antibiotics and 54 patients
in the 72-hour group received 24 hours of postoperative
antibiotics.
After 530 subjects had completed the primary outcome,
an external review assessed the futility based on the pre-
specified criterion using a conditional power of £20%.
Analysis revealed a conditional power of 0.12% under null
effects, 0.01% under continued trend effects, and 18% under
hypothesized effects, confirming little chance of achieving
significance. The study proceeded with enrollment during the
period of interim analysis and was terminated for futility
thereafter.
The study groups were well balanced with respect
to diagnoses and baseline clinical characteristics (Table I).
The study population had a mean age of 59 years and a mean
body mass index of 29 kg/m2, 16% of the study population
had diabetes, 23% of the study population were current
smokers, and most of the study population had degenerative
spondylolisthesis.
Surgical Treatment, Perioperative Details, and Postoperative
Outcomes
The type of surgical procedure, the number of surgical levels,
perioperative details, and postoperative outcomes were not
different between the 2 groups, except that patients in the 72-
hour group had a median 1-day longer hospital stay (Table III)
(p = 0.0001). All deep infections were treated with an irrigation
and debridement surgical procedure. The patient who had
adjacent-level discitis (organ or space surgical site infection)
was treated with inpatient admission and intravenous antibi-
otics and, at 3 years after the primary surgical procedure,
underwent a revision surgical procedure for pseudarthrosis and
kyphosis secondary to the infection.
Patient-Rated Health Outcomes
All patient-rated health outcome measures including the ODI,
leg and back pain, and physical and mental functioning improved
from the preoperative assessment to 1 year after the surgical
procedure (Table IV). The amount of improvement did not differ
between the 2 treatment groups for any outcome measure.

Infection Rates, Time to Infection, and Other Wound As-Treated Analysis

Complications
A complicated infection developed in 17 (6.0%) of 282 patients
in the 24-hour group and in 14 (5.2%) of 270 patients in the
72-hour group; the rates did not differ between antibiotic
groups (p = 0.714) (Table II). All complicated infections were
deep surgical site infections, except 1 in the 72-hour group that
was an organ or space surgical site infection. This patient had
adjacent-level discitis treated at 303 days after the initial sur-
gical procedure. The superficial infection rate was 9.6% (27 of
282) among patients in the 24-hour group and 8.1% (22 of
270) among patients in the 72-hour group and was not dif-
ferent between the groups (p = 0.654) (Table II). The median
time to infection was 15 days (range, 3 to 303 days). In the 24-
hour group, 8 (2.8%) of 282 patients developed an infection
that did not involve the surgical wound (i.e., urinary tract
infection or pneumonia). Similarly, in the 72-hour group, 11
(4.1%) of 270 patients developed a remote infection (p =
0.488). Other wound complications were not different between
the 2 groups (Table II).
An as-treated analysis based on the duration of antibiotics
received produced results similar to the primary analysis (Fig. 1).
The rate of complicated infection was 6.0% (20 of 333 patients) in
the 24-hour group and 5.0% (11 of 219 patients) in the 72-hour
group (p = 0.708). The rate of superficial infection was 10.2%
(34 of 333 patients) in the 24-hour group and 6.8% (15 of 219
patients) in the 72-hour group (p = 0.221).
Discussion
O ur randomized controlled trial of 552 patients undergoing
posterior thoracolumbar spinal surgery with placement
of a closed-suction drain found a similar rate of complicated
surgical site infection between patients randomized to 24 hours
and 72 hours of postoperative antibiotics. Furthermore, we
found no difference in the rate of superficial surgical site
infection. Our study was stopped after an interim analysis of
the data because a futility analysis showed that the results were
not likely to change with a large sample size. To further
establish this point, if we consider the overall infection rate in
 1738
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
the 2 treatment groups, a post hoc analysis with a power of 0.8
and an alpha of 0.05 suggests that the group sizes required to
detect a significant difference are upwards of 3,668 in each
group.
In our study, the drain was placed at the surgeon’s dis-
cretion. Although the use of drains remains common practice,
evidence for their use is controversial, with meta-analyses
showing no clear evidence supporting the use of postoperative
drains33. Rao et al. showed that the occurrence of surgical site
infection increased with increasing duration of drain use in a
case-control study of 1,587 patients undergoing arthrodesis10.
Similarly, Walsh et al. also demonstrated an increased risk of
surgical site infection in 5,473 patients undergoing arthrodesis
with postoperative drains11. However, several Level-II or lower-
evidence studies suggest that the use of a drain may not be a risk
factor for infection8,34-38.
The prolonged administration of antibiotics for the
duration of drain usage has been shown to reduce surgical site
infection after mastectomy and ventral hernia repair39,40.
However, drain use in these studies was 2 to 3 weeks, which is
much longer than that typically used in spinal surgery. In a
Level-I, prospective, randomized study of 314 patients under-
going thoracolumbar multilevel spinal surgery, Takemoto
et al.22 showed that extending postoperative antibiotics cover-
age until drain removal (mean, 3.2 days [range, 2 to 10 days])
did not affect the surgical site infection rate (13.2%) compared
with 24 hours of postoperative antibiotics (12.4%). A second
Level-II randomized controlled trial of 326 patients undergoing
arthrodesis showed a nonsignificant difference between anti-
biotics administered for 24 or 72 hours from the time of the
preoperative dose in patients with a closed-suction drain that
was removed after 48 hours41. That study, which was not
inclusive of lumbar surgical procedures and had a much lower
mean subject age (46 years), demonstrated a very low overall
infection rate of only 1.8% and was not generalizable to our
population with increased infection risk. Four other compar-
ative studies have also demonstrated no difference in surgical
site infection rates with the various prolonged postoperative
antibiotic regimes42-45.
Our study is in agreement with the infection rate re-
ported by Takemoto et al.22 (12.7% compared with 14.5% in
our study) and is in line with the infection rate associated with
similar higher-risk procedures1-4. Our patient cohort was likely
at an increased risk of infection as it excluded simple discec-
tomies or decompressions and minimally invasive surgical
approaches. Additionally, we used a prospective postoperative
surveillance method, which may have captured infections that
would otherwise have been missed in a retrospective analysis3,29.
It has been demonstrated that adherence to prospective data
collection is important to avoid underestimation of the oc-
currence of postoperative complications29. We also included
1 case of adjacent-level discitis. It is possible that this infection
was not related to the primary surgical procedure; however,
because it occurred within our observation time and we could
not definitively exclude involvement of the primary level, we
included it.
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N
In the present study, patients in the 2 antibiotic treatment
groups were well balanced in terms of patient-specific risk
factors for infection including smoking, diabetes, obesity, and a
revision surgical procedure4,46-48. Patients in the 72-hour group
had a longer hospital stay, indicating that longer antibiotic
administration increased the duration of hospital stay by 1 day.
However, importantly, at 1 year, patient-rated health pain and
physical functioning were not different between the 2 treat-
ment groups, indicating that prolonged administration has no
impact on patient functional recovery.
This study has a number of strengths. The study used a
prospective, randomized controlled design and has the largest
sample size of the 3 randomized controlled trials studying this
question. Additional prophylactic strategies were clearly de-
fined, and the eligibility criteria were quite strict to ensure a
homogeneous population relevant to patients undergoing
posterior thoracolumbar spinal surgery and having placement
of a closed-suction drain.
The limitations of our study included that our trial was
performed at a single center, which may have limited the
generalizability of the study. It was halted early and did not
complete the full enrollment deemed necessary in the a priori
power calculation. This raised the possibility that the study was
underpowered to detect differences, including differences in
the as-treated sensitivity test. However, futility analysis con-
firmed that the results were not likely to change, even with a
larger sample size. Criteria for drain insertion were not stan-
dardized, leading to a lack of clarity on indications and the
potential for an unrecognized difference between groups,
although the demographic characteristics, diagnoses, and pro-
cedures were similar. Physicians did not administer the anti-
biotics and were not blinded. For the primary analysis, patients
were included according to randomization regardless of the
treatment actually received, in compliance with the intention-
to-treat analysis; this likely underestimated the treatment effect.
Drain output and duration of drain use were not documented
because our standard protocol was to remove all drains by
postoperative day 2. The study was designed to ensure that all
patients in the experimental group received a minimum of 24
hours of antibiotics after the drain removal.
In conclusion, our work adds to the growing body of lit-
erature that suggests that the extension of postoperative anti-
biotics for 24 hours after the drain removal is not associated
with a reduction in the rate of complicated surgical site in-
fection after posterior thoracolumbar elective spinal surgery. n
Jennifer C. Urquhart, PhD1,2
Darryl Collings, MD3
Lori Nutt, RN1
Linda Kuska, RN1
Kevin R. Gurr, MD, FRCSC1,2,3
Fawaz Siddiqi, MD, FRCSC1,2,3
Parham Rasoulinejad, MD, FRCSC1,2,3
Alyssa Fleming2
 1739
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
Joanne Collie2
Christopher S. Bailey, MD, MSc, FRCSC1,2,3
1Combined Orthopaedic and Neurosurgical Spine Program, London
Health Sciences Centre, London, Ontario, Canada
2Lawson Health Research Institute, London, Ontario, Canada
3Division of Orthopaedics, Department of Surgery, Schulich School of Medicine
and Dentistry, University of Western Ontario, London, Ontario, Canada
References
1. Anderson PA, Savage JW, Vaccaro AR, Radcliff K, Arnold PM, Lawrence BD,
Shamji MF. Prevention of surgical site infection in spine surgery. Neurosurgery. 2017
Mar 1;80(3S):S114-23.
2. McClelland S 3rd, Takemoto RC, Lonner BS, Andres TM, Park JJ, Ricart-Hoffiz PA,
Bendo JA, Goldstein JA, Spivak JM, Errico TJ. Analysis of postoperative thoraco-
lumbar spine infections in a prospective randomized controlled trial using the Cen-
ters for Disease Control surgical site infection criteria. Int J Spine Surg. 2016 Apr 21;
10:14.
3. Roberts FJ, Walsh A, Wing P, Dvorak M, Schweigel J. The influence of surveillance
methods on surgical wound infection rates in a tertiary care spinal surgery service.
Spine (Phila Pa 1976). 1998 Feb 1;23(3):366-70.
4. Massie JB, Heller JG, Abitbol JJ, McPherson D, Garfin SR. Postoperative posterior
spinal wound infections. Clin Orthop Relat Res. 1992 Nov;284:99-108.
5. Khanna K, Janghala A, Sing D, Vail B, Arutyunyan G, Tay B, Deviren V. An analysis
of implant retention and antibiotic suppression in instrumented spine infections: a
preliminary data set of 67 patients. Int J Spine Surg. 2018 Aug 31;12(4):490-7.
6. Whitehouse JD, Friedman ND, Kirkland KB, Richardson WJ, Sexton DJ. The
impact of surgical-site infections following orthopedic surgery at a community hos-
pital and a university hospital: adverse quality of life, excess length of stay, and extra
cost. Infect Control Hosp Epidemiol. 2002 Apr;23(4):183-9.
7. Mirzai H, Eminoglu M, Orguc S. Are drains useful for lumbar disc surgery? A
prospective, randomized clinical study. J Spinal Disord Tech. 2006 May;19(3):
171-7.
8. Andrew Glennie R, Dea N, Street JT. Dressings and drains in posterior spine
surgery and their effect on wound complications. J Clin Neurosci. 2015 Jul;22(7):
1081-7. Epub 2015 Mar 25.
9. Waly F, Alzahrani MM, Abduljabbar FH, Landry T, Ouellet J, Moran K, Dettori JR.
The outcome of using closed suction wound drains in patients undergoing lumbar
spine surgery: a systematic review. Global Spine J. 2015 Dec;5(6):479-85.
10. Rao SB, Vasquez G, Harrop J, Maltenfort M, Stein N, Kaliyadan G, Klibert F,
Epstein R, Sharan A, Vaccaro A, Flomenberg P. Risk factors for surgical site infec-
tions following spinal fusion procedures: a case-control study. Clin Infect Dis. 2011
Oct;53(7):686-92.
11. Walsh TL, Querry AM, McCool S, Galdys AL, Shutt KA, Saul MI, Muto CA. Risk
factors for surgical site infections following neurosurgical spinal fusion operations: a
case control study. Infect Control Hosp Epidemiol. 2017 Mar;38(3):340-7. Epub
2016 Dec 19.
12. Sankar B, Ray P, Rai J. Suction drain tip culture in orthopaedic surgery: a pro-
spective study of 214 clean operations. Int Orthop. 2004 Oct;28(5):311-4. Epub
2004 Aug 14.
13. Olsen MA, Nepple JJ, Riew KD, Lenke LG, Bridwell KH, Mayfield J, Fraser VJ. Risk
factors for surgical site infection following orthopaedic spinal operations. J Bone
Joint Surg Am. 2008 Jan;90(1):62-9.
14. Allegranzi B, Zayed B, Bischoff P, Kubilay NZ, de Jonge S, de Vries F, Gomes SM,
Gans S, Wallert ED, Wu X, Abbas M, Boermeester MA, Dellinger EP, Egger M,
Gastmeier P, Guirao X, Ren J, Pittet D, Solomkin JS; WHO Guidelines Development
Group. New WHO recommendations on intraoperative and postoperative measures
for surgical site infection prevention: an evidence-based global perspective. Lancet
Infect Dis. 2016 Dec;16(12):e288-303. Epub 2016 Nov 2.
15. Bratzler DW, Houck PM; Surgical Infection Prevention Guideline Writers Work-
group. Antimicrobial prophylaxis for surgery: an advisory statement from the National
Surgical Infection Prevention Project. Am J Surg. 2005 Apr;189(4):395-404.
16. Shaffer WO, Baisden JL, Fernand R, Matz PG; North American Spine Society. An
evidence-based clinical guideline for antibiotic prophylaxis in spine surgery. Spine J.
2013 Oct;13(10):1387-92. Epub 2013 Aug 27.
17. Ban KA, Minei JP, Laronga C, Harbrecht BG, Jensen EH, Fry DE, Itani KM,
Dellinger EP, Ko CY, Duane TM. American College of Surgeons and Surgical Infection
Society: surgical site infection guidelines, 2016 update. J Am Coll Surg. 2017 Jan;
224(1):59-74. Epub 2016 Nov 30.
18. Barker FG 2nd. Efficacy of prophylactic antibiotic therapy in spinal surgery:
a meta-analysis. Neurosurgery. 2002 Aug;51(2):391-400, discussion:
400-1.
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N
ORCID iD for J.C. Urquhart: 0000-0002-9393-814X
ORCID iD for D. Collings: 0000-0003-1932-0285
ORCID iD for L. Nutt: 0000-0003-3836-0047
ORCID iD for L. Kuska: 0000-0001-5321-7447
ORCID iD for K.R. Gurr: 0000-0003-4348-268X
ORCID iD for F. Siddiqi: 0000-0002-9247-6887
ORCID iD for P. Rasoulinejad: 0000-0002-3031-2633
ORCID iD for A. Fleming: 0000-0002-1888-1796
ORCID iD for J. Collie: 0000-0003-0231-6757
ORCID iD for C.S. Bailey: 0000-0001-5508-8415
19. Hellbusch LC, Helzer-Julin M, Doran SE, Leibrock LG, Long DJ, Puccioni MJ,
Thorell WE, Treves JS. Single-dose vs multiple-dose antibiotic prophylaxis in in-
strumented lumbar fusion—a prospective study. Surg Neurol. 2008 Dec;70(6):
622-7; discussion 627. Epub 2008 Jan 18.
20. Dimick JB, Lipsett PA, Kostuik JP. Spine update: antimicrobial prophylaxis in
spine surgery: basic principles and recent advances. Spine (Phila Pa 1976). 2000
Oct 1;25(19):2544-8.
21. Prokuski L. Prophylactic antibiotics in orthopaedic surgery. J Am Acad Orthop
Surg. 2008 May;16(5):283-93.
22. Takemoto RC, Lonner B, Andres T, Park J, Ricart-Hoffiz P, Bendo J, Goldstein J,
Spivak J, Errico T. Appropriateness of twenty-four-hour antibiotic prophylaxis after
spinal surgery in which a drain is utilized: a prospective randomized study. J Bone
Joint Surg Am. 2015 Jun 17;97(12):979-86.
23. Balch A, Wendelboe AM, Vesely SK, Bratzler DW. Antibiotic prophylaxis for
surgical site infections as a risk factor for infection with Clostridium difficile. PLoS
One. 2017 Jun 16;12(6):e0179117.
24. Bryson DJ, Morris DL, Shivji FS, Rollins KR, Snape S, Ollivere BJ. Antibiotic
prophylaxis in orthopaedic surgery: difficult decisions in an era of evolving antibiotic
resistance. Bone Joint J. 2016 Aug;98-B(8):1014-9.
25. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health
care-associated infection and criteria for specific types of infections in the acute care
setting. Am J Infect Control. 2008 Jun;36(5):309-32.
26. DeVine J, Norvell DC, Ecker E, Fourney DR, Vaccaro A, Wang J, Andersson G.
Evaluating the correlation and responsiveness of patient-reported pain with function
and quality-of-life outcomes after spine surgery. Spine (Phila Pa 1976). 2011 Oct 1;
36(21)(Suppl):S69-74.
27. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976).
2000 Nov 15;25(22):2940-52; discussion 2952.
28. Lee CE, Browell LM, Jones DL. Measuring health in patients with cervical and
lumbosacral spinal disorders: is the 12-item short-form health survey a valid alter-
native for the 36-item short-form health survey? Arch Phys Med Rehabil. 2008 May;
89(5):829-33.
29. Street JT, Lenehan BJ, DiPaola CP, Boyd MD, Kwon BK, Paquette SJ, Dvorak MF,
Rampersaud YR, Fisher CG. Morbidity and mortality of major adult spinal surgery. A
prospective cohort analysis of 942 consecutive patients. Spine J. 2012 Jan;12(1):
22-34. Epub 2011 Dec 29.
30. Lachin JM. A review of methods for futility stopping based on conditional power.
Stat Med. 2005 Sep 30;24(18):2747-64.
31. Schulz KF, Grimes DA. Multiplicity in randomised trials II: subgroup and interim
analyses. Lancet. 2005 May 7-13;365(9471):1657-61.
32. Snapinn S, Chen MG, Jiang Q, Koutsoukos T. Assessment of futility in clinical
trials. Pharm Stat. 2006 Oct-Dec;5(4):273-81.
33. Patel SB, Griffiths-Jones W, Jones CS, Samartzis D, Clarke AJ, Khan S,
Stokes OM. The current state of the evidence for the use of drains in spinal
surgery: systematic review. Eur Spine J. 2017 Nov;26(11):2729-38. Epub 2017
Feb 11.
34. Adogwa O, Elsamadicy AA, Sergesketter AR, Shammas RL, Vatsia S, Vuong VD,
Khalid S, Cheng J, Bagley CA, Karikari IO. Post-operative drain use in patients
undergoing decompression and fusion: incidence of complications and symptomatic
hematoma. J Spine Surg. 2018 Jun;4(2):220-6.
35. Payne DH, Fischgrund JS, Herkowitz HN, Barry RL, Kurz LT, Montgomery DM.
Efficacy of closed wound suction drainage after single-level lumbar laminectomy. J
Spinal Disord. 1996 Oct;9(5):401-3.
36. 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 Oct;468(10):2690-4. Epub 2010 Jan 21.
37. Yao R, Tan T, Tee JW, Street J. Prophylaxis of surgical site infection in adult spine
surgery: a systematic review. J Clin Neurosci. 2018 Jun;52:5-25. Epub 2018 Mar 30.
38. Brown MD, Brookfield KF. A randomized study of closed wound suction drainage
for extensive lumbar spine surgery. Spine (Phila Pa 1976). 2004 May 15;29(10):
1066-8.
 1740
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 101 -A N U M B E R 19 O C T O B E R 2, 2 019
d d
39. Edwards BL, Stukenborg GJ, Brenin DR, Schroen AT. Use of prophylactic post-
operative antibiotics during surgical drain presence following mastectomy. Ann Surg
Oncol. 2014 Oct;21(10):3249-55. Epub 2014 Aug 20.
40. Wong A, Lee S, Nathan NS, Wang F, Hansen SL, Harris HW, Sbitany H. Post-
operative prophylactic antibiotic use following ventral hernia repair with placement of
surgical drains reduces the postoperative surgical-site infection rate. Plast Reconstr
Surg. 2016 Jan;137(1):285-94.
41. Marimuthu C, Abraham VT, Ravichandran M, Achimuthu R. Antimicrobial
prophylaxis in instrumented spinal fusion surgery: a comparative analysis of
24-hour and 72-hour dosages. Asian Spine J. 2016 Dec;10(6):1018-22. Epub 2016
Dec 8.
42. Kim B, Moon SH, Moon ES, Kim HS, Park JO, Cho IJ, Lee HM. Antibiotic microbial
prophylaxis for spinal surgery: comparison between 48 and 72-hour amp protocols.
Asian Spine J. 2010 Dec;4(2):71-6. Epub 2010 Nov 24.
43. Takahashi H, Wada A, Iida Y, Yokoyama Y, Katori S, Hasegawa K, Shintaro T,
Suguro T. Antimicrobial prophylaxis for spinal surgery. J Orthop Sci. 2009 Jan;14(1):
40-4. Epub 2009 Feb 13.
T H E E F F E C T O F P R O L O N G E D P O S T O P E R AT I V E A N T I B I O T I C
A D M I N I S T R AT I O N O N T H E R AT E O F I N F E C T I O N
44. Ohtori S, Inoue G, Koshi T, Yamashita M, Yamauchi K, Suzuki M, Orita S, Eguchi
Y, Ochiai N, Kishida S, Takaso M, Takahashi K. Long-term intravenous administra-
tion of antibiotics for lumbar spinal surgery prolongs the duration of hospital stay and
time to normalize body temperature after surgery. Spine (Phila Pa 1976). 2008 Dec
15;33(26):2935-7.
45. Lewis A, Lin J, James H, Hill TC, Sen R, Pacione D. Discontinuation of postoper-
ative prophylactic antibiotics after noninstrumented spinal surgery: results of a quality
improvement project. Neurohospitalist. 2018 Jul;8(3):129-34. Epub 2018 Jan 21.
46. Pesenti S, Pannu T, Andres-Bergos J, Lafage R, Smith JS, Glassman S, de
Kleuver M, Pellise F, Schwab F, Lafage V; Scoliosis Research Society (SRS). What
are the risk factors for surgical site infection after spinal fusion? A meta-analysis. Eur
Spine J. 2018 Oct;27(10):2469-80. Epub 2018 Aug 20.
47. Parchi PD, Evangelisti G, Andreani L, Girardi F, Darren L, Sama A, Lisanti M.
Postoperative spine infections. Orthop Rev (Pavia). 2015 Sep 28;7(3):5900.
48. Fei Q, Li J, Lin J, Li D, Wang B, Meng H, Wang Q, Su N, Yang Y. Risk factors for
surgical site infection after spinal surgery: a meta-analysis. World Neurosurg. 2016
Nov;95:507-15. Epub 2015 Jun 6.