A multicenter initiative to reduce intrathecal baclofen pump surgical site infection: a Cerebral Palsy Research Network quality improvement project

Robert J. Bollo Department of Neurosurgery, Division of Pediatric Neurosurgery, Primary Children’s Hospital, University of Utah, Salt Lake City, Utah;

Search for other papers by Robert J. Bollo in
jns
Google Scholar
PubMed
Close
 MD
,
Paul H. Gross Cerebral Palsy Research Network, Greensville, South Carolina;

Search for other papers by Paul H. Gross in
jns
Google Scholar
PubMed
Close
 BA
,
Brandon G. Rocque Department of Neurosurgery, Division of Pediatric Neurosurgery, Children’s Hospital of Alabama, University of Alabama School of Medicine, Birmingham, Alabama;

Search for other papers by Brandon G. Rocque in
jns
Google Scholar
PubMed
Close
 MD, MS
,
Samuel R. Browd Department of Neurosurgery, Division of Pediatric Neurosurgery, Seattle Children’s Hospital, University of Washington School of Medicine, Seattle, Washington;

Search for other papers by Samuel R. Browd in
jns
Google Scholar
PubMed
Close
 MD, PhD
,
Jeffrey S. Raskin Department of Neurosurgery, Division of Pediatric Neurosurgery, Riley Children’s Hospital, University of Indiana School of Medicine, Indianapolis, Indiana;

Search for other papers by Jeffrey S. Raskin in
jns
Google Scholar
PubMed
Close
 MD, MS
,
Jeffrey R. Leonard Department of Neurosurgery, Division of Pediatric Neurosurgery, Nationwide Children’s Hospital, The Ohio State College of Medicine, The Ohio State University, Columbus, Ohio;

Search for other papers by Jeffrey R. Leonard in
jns
Google Scholar
PubMed
Close
 MD
,
Lama Albarqawi Department of Population Health Sciences, University of Utah School of Medicine, Salt Lake City, Utah;

Search for other papers by Lama Albarqawi in
jns
Google Scholar
PubMed
Close
 MS
,
Amy F. Bailes Division of Occupational Therapy and Physical Therapy, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio; and
Department of Rehabilitation Exercise and Nutrition Sciences, University of Cincinnati School of Medicine, Cincinnati, Ohio

Search for other papers by Amy F. Bailes in
jns
Google Scholar
PubMed
Close
 PT, PhD
, and
for the Cerebral Palsy Research Network (CPRN)
Search for other papers by for the Cerebral Palsy Research Network (CPRN) in
jns
Google Scholar
PubMed
Close
for the Cerebral Palsy Research Network (CPRN)
Free access

OBJECTIVE

Intrathecal baclofen (ITB) therapy is an effective treatment for spasticity and dystonia in children with cerebral palsy (CP). However, ITB pump surgery is associated with one of the highest rates of surgical site infection (SSI) in medicine, leading to significant morbidity and expense. Surgical protocols have reduced the rate of SSI in children with other CNS implants, and single-center protocols have been effective in ITB surgery in pediatrics. The authors describe the first multicenter quality improvement (QI)–driven standardized protocol for ITB pump surgery in children with CP across the Cerebral Palsy Research Network (CPRN), implemented with the aim of reducing ITB-associated SSI.

METHODS

SSI was defined as a culture-positive infection, ITB pump system removal for suspected infection, or wound dehiscence with exposed hardware. Each center reported historical infection rates for at least 3 years before initiating the SSI protocol (preintervention phase). After initiation of a 13-step surgical protocol, a consecutive series of 130 patients undergoing 149 surgical procedures for ITB at four CPRN tertiary pediatric neurosurgery centers were prospectively enrolled at surgery during a 2-year study period (intervention phase). QI methodology was used, including development of a key driver diagram and tracking performance using run and control charts. The primary process measure goal was documented compliance with 80% of the protocol steps, and the primary outcome measure goal was a 20% reduction in 90-day infection rate. Patient characteristics were collected from the CPRN Research Electronic Data Capture registry, including age at surgery, BMI, Gross Motor Function Classification System level, and pattern of spasticity.

RESULTS

The aggregated preintervention 90-day ITB SSI rate was 4.9% (223 procedures) between 2014 and 2017. During the intervention phase, 136 of 149 ITB surgeries performed met inclusion criteria for analysis. The mean documented compliance rate with protocol steps was 75%, and the 90-day infection rate was 4.4%, with an average of 42 days from index surgery to infection.

CONCLUSIONS

This is the first multicenter QI initiative designed to reduce SSI in ITB surgery in children with CP. Ongoing enrollment and expansion of the protocol to other CPRN centers will facilitate identification of patient- and procedure-specific risk factors for SSI, and iterative plan-do-study-act cycles incorporating these data will further decrease the risk of SSI for ITB surgery in children.

ABBREVIATIONS

CP = cerebral palsy; CPRN = Cerebral Palsy Research Network; GMFCS = Gross Motor Function Classification System; HCRN = Hydrocephalus Clinical Research Network; ITB = intrathecal baclofen; IV = intravenous; PDSA = plan-do-study-act; QI = quality improvement; SMART = specific, measurable, actionable, realistic, and timely; SSI = surgical site infection.

OBJECTIVE

Intrathecal baclofen (ITB) therapy is an effective treatment for spasticity and dystonia in children with cerebral palsy (CP). However, ITB pump surgery is associated with one of the highest rates of surgical site infection (SSI) in medicine, leading to significant morbidity and expense. Surgical protocols have reduced the rate of SSI in children with other CNS implants, and single-center protocols have been effective in ITB surgery in pediatrics. The authors describe the first multicenter quality improvement (QI)–driven standardized protocol for ITB pump surgery in children with CP across the Cerebral Palsy Research Network (CPRN), implemented with the aim of reducing ITB-associated SSI.

METHODS

SSI was defined as a culture-positive infection, ITB pump system removal for suspected infection, or wound dehiscence with exposed hardware. Each center reported historical infection rates for at least 3 years before initiating the SSI protocol (preintervention phase). After initiation of a 13-step surgical protocol, a consecutive series of 130 patients undergoing 149 surgical procedures for ITB at four CPRN tertiary pediatric neurosurgery centers were prospectively enrolled at surgery during a 2-year study period (intervention phase). QI methodology was used, including development of a key driver diagram and tracking performance using run and control charts. The primary process measure goal was documented compliance with 80% of the protocol steps, and the primary outcome measure goal was a 20% reduction in 90-day infection rate. Patient characteristics were collected from the CPRN Research Electronic Data Capture registry, including age at surgery, BMI, Gross Motor Function Classification System level, and pattern of spasticity.

RESULTS

The aggregated preintervention 90-day ITB SSI rate was 4.9% (223 procedures) between 2014 and 2017. During the intervention phase, 136 of 149 ITB surgeries performed met inclusion criteria for analysis. The mean documented compliance rate with protocol steps was 75%, and the 90-day infection rate was 4.4%, with an average of 42 days from index surgery to infection.

CONCLUSIONS

This is the first multicenter QI initiative designed to reduce SSI in ITB surgery in children with CP. Ongoing enrollment and expansion of the protocol to other CPRN centers will facilitate identification of patient- and procedure-specific risk factors for SSI, and iterative plan-do-study-act cycles incorporating these data will further decrease the risk of SSI for ITB surgery in children.

In Brief

The rate of baclofen pump surgical site infection (SSI) in children is very high and represents a significant source of morbidity, yet the patient- and procedure-specific risk factors and effective strategies to mitigate infection risk are poorly understood. The authors conducted the first multicenter prospective effort to study this problem using the science of quality improvement through the Cerebral Palsy Research Network, demonstrating 75% compliance with a 13-step surgical protocol and an overall infection rate of 4.2%. Using these data to further refine a protocol and expanding its use may lead to significant reduction of SSI complicating CSF shunt surgery.

Cerebral palsy (CP) is a group of mobility disorders common in childhood1 that are typically caused by injury to the developing brain or abnormal brain development. The primary disorder is spasticity.2 The use of intrathecal baclofen (ITB) therapy to treat spasticity was initially described for adults with spinal cord injury, but its use for patients with CP was subsequently described.3 The use of ITB for children with spasticity, dystonia, and mixed hypertonic disorders in the context of CP has become increasingly common in clinical practice.

A growing body of literature has recognized a variable but high rate of surgical site infections (SSIs) associated with the use of ITB in pediatric patients worldwide. Reported infection rates in single-institution retrospective studies vary between 3% and 40%, with a mean infection rate of approximately 10%.413 Single-center retrospective analyses have reported patient-specific risk factors for infection in children, including low BMI,6,9 gastrostomy tube placement,6,9 incontinence,6 mixed movement disorder,9 epilepsy,9 scoliosis,13 young age,13 and American Society of Anesthesiologists class III or greater,10 as well as procedure-specific factors, such as subfascial placement,11 wound dehiscence,13 CSF leak,9 and number of surgical revisions.13

The definition and clinical management of ITB-associated SSI reported in the literature are highly variable,413 obfuscating our understanding of the problem, but its impact on children with CP is profound. SSI typically requires surgery for device explantation, and pump removal is often complicated by both baclofen withdrawal and meningitis. Furthermore, a large proportion of children with CP managed using ITB have concurrent hydrocephalus and CSF shunts, which in the context of CSF infection may also require shunt removal and intravenous (IV) antimicrobial treatment in the hospital with external CSF diversion, followed by surgery for pump reimplantation. A recent single-center quality improvement (QI) initiative demonstrated promising results after applying QI methodology to ITB surgery in children;5 however, to date, patient- and procedure-specific risk factors have not been validated, and the time frame during which patients remain at risk for ITB-associated SSI has not been characterized.

QI methods hold significant potential to address this problem. Because SSI is also a significant clinical problem in children with hydrocephalus treated with CSF shunts, the multicenter Hydrocephalus Clinical Research Network (HCRN) has used QI methodology to successfully identify risk factors for shunt infection and surgical protocols associated with a significant reduction in infection rate.14,15 This experience highlights the potential benefits that may be achieved by applying QI methodology to ITB-associated SSI in pediatric patients in a multicenter network.

Furthermore, learning health networks have been used for other pediatric conditions to enable multiple centers to work together and apply QI strategies to reduce variation in care and improve clinical outcomes.16 The Cerebral Palsy Research Network (CPRN) was established in 2015 across multiple institutions as a means to identify problems, generate solutions, and answer questions that cannot be solved at a single institution in patients with CP because of the clinical variability in the population and the longitudinal course of the disease.17 Here, we report the first multicenter QI initiative within the CPRN to characterize, monitor, and reduce ITB-associated SSI in children with CP.

Methods

Setting and Context

The intervention phase of the QI project began in January 2018 and is reported over a 3-year period until December 2020 across four pediatric academic centers participating in the CPRN (Primary Children’s Hospital, Children’s Hospital of Alabama, Seattle Children’s Hospital, and Riley Children’s Hospital). Each center’s IRB reviewed the project and determined that no IRB oversight or informed consent was necessary. The QI team consisted of pediatric neurosurgeons from each participating CPRN center, an expert trained and experienced in QI methodology, and the parent of a child with CP. All neurosurgeons on the QI team were familiar with the HCRN protocol to prevent infection in hydrocephalus14,15 as well as previous single-center efforts to reduce ITB infection.5 The team met twice each month to review information, evaluate progress, make decisions, and plan next steps. Ongoing team education on improvement science and coaching supported the implementation and reporting of the project.

The “model for improvement” framework was used.18 This framework promotes increased effectiveness, efficiency, and learning and was used to guide planning and implementation of improvement efforts. In addition to the plan-do-study-act (PDSA) cycle,19 the model for improvement is based on 3 questions: 1) what are we trying to accomplish; 2) how will we know that change is an improvement; and 3) what changes can we make that will result in improvement? The model for improvement was used to help the team establish a shared vision for study goals using global and SMART (specific, measurable, actionable, realistic, and timely) aims, formulate a measurement system to identify specific changes resulting in improvement, and design interventions that could result in the desired improvement. Both a process measure (what the provider experiences) and an outcome measure (what the patient experiences) were defined by the team.

Process Measure and Outcome Measure Definitions

The process measure for this study was the percentage of infection reduction protocol steps documented by the surgical team. The goal was to increase the percentage to 80% by December 2020. The team selected this process measure rather than the percentage of procedures compliant with all protocol steps because of variable access to some protocol steps (e.g., antibiotic sutures) at participating institutions.

The outcome measure was the infection rate in individuals with CP undergoing ITB pump procedures. The objective was to decrease the 90-day infection rate by 20% from baseline by December 2020.

The QI team operationally defined infection at the time of a surgical procedure as system explant for suspected infection, culture-positive infection, or wound dehiscence with exposed hardware. When any of these were selected and the infection reduction protocol form was completed for the procedure, infection was assumed. If a surgical procedure was not performed, the patient did not meet criteria for infection. For the outcome measure, only infections that occurred within 90 days were considered. However, because the time frame during which a patient undergoing surgery for ITB is at risk for an SSI is unknown, we continued to collect data on late infections occurring after this time frame if the index procedure was recorded in the CPRN registry. Although most patients were pediatric, the population was defined as patients undergoing any ITB-related surgery with a diagnosis of CP, without a specific age limit.

The team developed a key driver diagram to represent the theory of improvement and to guide overall planning of interventions (Fig. 1). The key driver diagram was iteratively refined and revised as the project progressed. Change concepts were used to brainstorm ideas for potential interventions that could result in an improvement in the process and outcome measures.18

FIG. 1.
FIG. 1.

Key driver diagram representing project to increase the number of protocol steps documented as completed in the operating room (OR) and to reduce infection rate. REDCap = Research Electronic Data Capture.

Preintervention Phase

Before interventions were initiated, a retrospective chart review was performed to evaluate the annual baseline infection rate from 2014 through 2017 across the four centers. Because it was not possible to separate patients with CP from those with other diagnoses, all ITB pump procedures were included for the calculation of the baseline infection rate. Although each center reportedly performed some procedural steps to mitigate infection risk in the preintervention phase, a standardized protocol was not in place or performed at any center. In addition, it was not possible to verify that the same definition agreed upon by the team and used in the intervention phase was identical in all cases across all centers to hospital-based definitions of SSI used retrospectively in the preintervention phase to establish the baseline SSI rate.

Intervention Phase

The surgeon or designated member of the operative team was responsible for completing the infection reduction protocol form for each procedure. At three centers, data were collected in real time by a research coordinator or nurse and submitted to the data coordination center at the University of Utah after each ITB surgical procedure. One center collected and submitted data retrospectively after each case. This was completed as soon as possible (within a few days) after surgery, the data were verified against the medical and operative records, and any concerns were discussed with the surgeon. Data were confirmed by the data coordination center and assessed for completeness and accuracy by the study team on a quarterly basis. The percentage of infection reduction protocol steps followed was plotted on a run chart quarterly during the intervention phase. Infection rate was plotted on a run chart annually in both the preintervention and intervention phases. Surgical procedures were excluded if a pump was removed without replacement for another cause without evidence of infection, as this did not result in an ongoing risk of device-related SSI.

Intervention

We achieved consensus on a protocol by incorporating steps taken at each site to reduce SSI in ITB pump surgery in children with the ITB pump protocol reported by Desai et al.5 and modeling the intervention after the HCRN shunt infection protocol.14,15 Consensus on the final 13-step protocol (Fig. 2) was achieved by PDSA cycles19 using the model for improvement framework,18 including brainstorming change concepts using key driver diagrams as described above.

FIG. 2.
FIG. 2.

Infection reduction protocol form. ABX = antibiotics; CHG = chlorhexidine gluconate; Gent = gentamicin; Vanc = vancomycin. © Cerebral Palsy Research Network, published with permission.

Data Analysis

The process was evaluated through a quantitative time series study design. Statistical process control methods were used to monitor changes in the process and outcome measures. Annotated control charts representing the process and outcome measures were updated based on the reports from the participating institutions and shared with the QI core team quarterly.20 The centerline represented the mean with upper and lower control limits, which are set by convention at ± 3 standard deviations beyond the mean.18,20 Standard criteria were used to determine whether observed changes in measures were common cause variation or due to a special cause, in this case the intervention. These criteria include observation of 8 points above or below the centerline indicating a change in the measure.18 We relied on standard statistical process control methods to define rules for detecting special cause variation incorporating alpha and beta error, so power calculations were not performed.21

Results

Preintervention Period

During the 4-year period covered by the retrospective data collection, 223 pump surgeries were performed and 11 infections (4.9%) occurred. Most patients were treated for spasticity or mixed spasticity and dystonia (data not shown).

Intervention Period

During the study period from January 1, 2018, through December 31, 2020, 149 ITB-related surgical procedures were performed on 130 patients. These procedures included 65 new systems, 57 pump replacements, 11 catheter revisions, 2 wound revisions, 13 system removals, and 1 pump repositioning. The 13 system removals that did not result in replacement pumps at risk for SSI were excluded, resulting in 136 procedures that could be assessed for infection rate. The mean (SD) patient age at surgery was 12.1 (4.8) years, the mean BMI was 17.1 (4.0), and the majority of patients (n = 104, 76%) were classified as Gross Motor Function Classification System (GMFCS) level IV or V at the time of surgery (Table 1). The majority of patients were treated for spasticity or mixed spasticity and dystonia.

TABLE 1.

Characteristics of patients undergoing 136 pump procedures during the intervention period

VariableAll Procedures, n = 136No Infection, n = 130Infection, n = 6
Mean age (SD), yrs12.1 (4.8)13.2 (4.9)12.2 (7.5)
Sex, n
 Female48480
 Male88826
Race, n
 American Indian or Alaska Native330
 Asian330
 Black or African American11101
 Native Hawaiian or other Pacific Islander110
 White1061015
 Multiple220
 Unknown10100
Ethnicity, n
 Hispanic18180
 Not Hispanic1121066
 Unknown660
Mean BMI (SD)17.1 (4.0)17.2 (4.0)16.3 (3.2)
BMI not reported, n40400
GMFCS level, n
 I110
 II10100
 III862
 IV32302
 V72702
 Not reported13130
Movement disorder, n
 Spasticity79736
 Dystonia110
 Dystonia110
 Mixed56560
Index surgery type, n
 New system65623
 Pump replacement57552
 Catheter revision11101
 Wound revisions220
 Repositioning110
CP distribution, n
 Unilat220
 Bilat symmetric101956
 Bilat asymmetric550
 Not reported28280

Process Measure: Documentation of Completed Steps

Figure 3 displays the percentage of infection reduction protocol steps documented quarterly across time; the percentage of steps completed ranged from 66% (3rd quarter 2019) to 81% (4th quarter 2020). Overall, across all procedures during the reported study period, a mean of 9.7 steps were completed (75%, range 3–12 steps). The process measure goal of 80% was not met, although a 75% compliance rate was. No patient had all 13 steps completed.

FIG. 3.
FIG. 3.

Mean percentage of infection reduction protocol steps documented as completed over time. Figure is available in color online only.

Table 2 displays the frequency with which each infection reduction protocol step was completed. The most frequently completed steps were step 2 (IV antibiotics 60 minutes before procedure, 94.1%) and step 8 (antimicrobial irrigation, 94.1%). The least frequently completed steps were step 11 (antibiotic sutures, 27.9%) and step 12 (antibiotic powder in wounds, 45.6%). Five steps were completed at least 90% of the time (Table 2).

TABLE 2.

Steps and number of infection reduction protocol procedures that completed each step

Step No.Protocol StepNo. (%) Completed
1CHG wipes106 (77.9)
2IV antibiotics w/in 60 mins 128 (94.1)
3Preop infection reduction 127 (93.4)
4Chloraprep 127 (93.4)
5Ioban 107 (78.7)
6Water-based scrub 97 (71.3)
7Double glove 100 (73.5)
8Antimicrobial irrigation 128 (94.1)
9Antimicrobial soak/wrap 94 (69.1)
10Intrathecal antibiotics 102 (75.0)
11Antibiotic sutures 38 (27.9)
12Antibiotic powder in wounds62 (45.6)
13Postop antibiotics126 (92.6)

CHG = chlorhexidine gluconate.

Outcome Measure: 90-Day SSI Rate

The 90-day infection rate was plotted over time (Fig. 4). During the preimplementation period, the infection rate was 4.9%. After the protocol was implemented, 6 procedures were complicated by an SSI within 90 days, yielding a 4.4% infection rate. Since implementation, the data do not show a change in the measure or shift of the centerline, indicating that the outcome measure goal was not met. In the 6 patients who experienced infection after the protocol was implemented (Table 3), the mean number of days to infection, measured as days to pump procedure to address infection, was 42 (range 33–55) days. In addition, 3 patients experienced infection at more than 90 days after the pump procedure (Table 4).

FIG. 4.
FIG. 4.

Percentage of pump infections within 30 days (y-axis) over time (x-axis). The preimplementation period included 2014–2017, and the postimplementation period was initiated in January 2018. No special cause variation was observed in the data. Figure is available in color online only.

TABLE 3.

Characteristics of patients with infection < 90 days after pump procedure

Pt No.Age (yrs), SexRaceBMIGMFCS LevelIndex SurgeryNo. (%) of Steps CompletedMissing StepsDays to InfectionReason for Surgery
1 25, MWhite 19.82VPump replacement11 (85)11, 1233Culture positive
2 9, MWhite 16.07VCatheter revision10 (77)10–1235Culture positive
3 6, MWhite 11.89IIINew pump11 (85)11, 1241Culture positive
4 16, MBlack 20.22IVPump replacement10 (77)10, 12, 1339Culture positive
5 5, MWhite 14.25IVNew pump11 (85)11, 1255Culture positive
6 13, MWhite 16.81IIINew pump12 (92)1149Suspected infection

Pt = patient.

None of the patients were Hispanic, all CP distribution was bilateral symmetric, all patients had spasticity movement disorder, and all patients underwent system removal at the time of infection.

TABLE 4.

Characteristics of patients with infection > 90 days after pump procedure

Pt No.Age (yrs), SexRaceBMIGMFCS LevelIndex SurgeryNo. (%) of Steps CompletedMissing StepsTime to InfectionReason for SurgerySurgery Type at Time of Infection
713, MWhite15.54IIPump replacement8 (62)9–132 yrsCulture positive System removal
86, M>116.2IVNew pump placement8 (62)5–7, 9, 11245 daysSuspected infectionSystem removal
918, MMissing23.0MissingPump replacement11 (85)6, 7196 daysSuspected infectionWound revision

None of the patients were Hispanic, all CP distribution was bilateral symmetric, and all patients had spasticity movement disorder.

Discussion

Surgery to implant baclofen pumps in children has a very high SSI rate.413 Previous work by Desai and colleagues5 demonstrated a significant reduction in the SSI rate at a single center through the implementation of a surgical protocol. Ongoing QI work by the HCRN has demonstrated a significant reduction of SSI complicating CSF shunt surgery across a network, with excellent compliance. This is the first multicenter network-based QI initiative to minimize the SSI rate for ITB pump surgery in children with CP through adaptation of a surgical protocol.

Compared with single-center studies, network-based protocols are more likely to have higher variation and lower compliance rates.14,15 On the basis of the work of Desai et al.,5 in which an 82% compliance rate was reported, we set a process measure goal of 80% compliance that we failed to meet. However, our compliance rate of 75% compares favorably with other network-based surgical protocols.14,15,22 Two protocol steps, antibiotic sutures and vancomycin powder, were not available during the study period at one center. These were the most commonly missed protocol steps, and 4 of 6 infections were not compliant with these 2 steps. In addition, the compliance rate improved throughout the study, from 66% at the outset to 81% at the conclusion of the study period.

For the outcome measure, we sought to move the centerline to demonstrate a decreased postimplantation infection rate. Although we reduced the infection rate from 4.9% before implementation to 4.4% postimplantation, we did not meet our goal of shifting the centerline. This may be attributable to the low baseline infection rate, where a significant change requires more power to detect. In addition, all four centers are familiar with the published HCRN CSF shunt protocol,14,15 and three are participating HCRN centers; this emphasis on infection reduction could also have contributed to a lower baseline infection rate. Additionally, surgical protocols to reduce SSI in pediatric neurosurgery are becoming more commonplace, and these factors may all have contributed to a low baseline infection rate.2325 Finally, the SSI definition was applied only postintervention and may be broader than institution-specific SSI definitions reported in the preintervention period.

This initial work comprises data across four CPRN centers. To improve the power of our prospective QI initiative and detect clinically and statistically significant relationships between our process (compliance) and outcome (SSI) measures, as well as to evaluate the impact of compliance with individual protocol steps on the outcome measure, we will include additional patients and centers in this ongoing prospective QI initiative. Thus, we have created a formal change package available within the CPRN and are actively recruiting other centers to join this initiative. Furthermore, the highest-risk time frame for SSI after baclofen pump surgery in children is not well defined, and while our primary outcome measure is the 90-day infection rate, we are also collecting types of infections that occur beyond 90 days to better characterize the temporal risk (Table 3). Finally, patient-specific factors such as age, BMI, GMFCS level, dystonia, and complex chronic conditions and their relationship to infection are poorly defined for baclofen pump surgery. The CPRN collects patient-specific factors throughout the network, and we hope to characterize modifiable patient-specific factors that convey a high risk for SSI to further reduce infection.

Limitations of the Study

This study was limited by the relatively small number of ITB procedures identified during the study period, preventing a quantitative analysis of infection risk. By adding study sites, we aim to mitigate this problem in future assessments. In addition, the fidelity of data in the preintervention cohort was significantly lower compared with that of the postintervention cohort. For the preintervention group, in which procedures were not captured in the CPRN registry, infection data were derived from institutional databases unique to each center. We also do not have detailed demographic, diagnostic, procedural, or other clinical information available for this cohort. Also, although it is standard practice to explant the ITB system in the context of SSI at all study centers, we only captured surgical data, so it is possible (although unlikely) an SSI was treated successfully without surgery and therefore not captured. Finally, we did not capture data regarding SSI bacterial pathogens identified in patients with culture-positive infections. However, this will be added as a data element moving forward for the postintervention group.

Conclusions

The broad purpose of this initiative was to characterize, monitor, and reduce ITB SSI in children with CP. Although our initial findings did not demonstrate a significant reduction in SSI rates with this protocol, this ongoing prospective, network-based QI initiative continues to expand and accumulate data and as such may represent a powerful tool to decrease a significant source of morbidity in children with CP.

Acknowledgments

We thank Kristin Kraus for editorial assistance. The study was partially funded by the Schwab Charitable Foundation.

Disclosures

Mr. Gross reported a nonfinancial relationship with the Cerebral Palsy Research Network, which receives funding from Neurocrine Biosciences, PTC Therapeutics, TEVA Pharmaceuticals, and Rifton. Dr. Raskin reported receiving honoria from Medtronic.

Author Contributions

Conception and design: Bollo, Bailes. Acquisition of data: all authors. Analysis and interpretation of data: all authors. Drafting the article: Bollo, Bailes. Critically revising the article: all authors. Reviewed submitted version of manuscript: Bollo. Approved the final version of the manuscript on behalf of all authors: Bollo. Administrative/technical/material support: Gross. Study supervision: Bollo, Bailes.

Supplemental Information

Previous Presentations

A subset of the data included in this paper was presented as a poster at the Combined American Academy for Cerebral Palsy and Developmental Medicine 73rd Annual and International Alliance of Academies of Childhood Disability 2nd Triannual Meeting in Anaheim, California, September 18–21, 2019.

References

  • 1

    What is cerebral palsy? Centers for Disease Control and Prevention. US Department of Health and Human Services. May 2, 2022. Accessed January 24, 2023. https://www.cdc.gov/ncbddd/cp/facts.html#:~:text=Cerebral%20palsy%20(CP)%20is%20a,problems%20with%20using%20the%20muscles

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Hasnat MJ, Rice JE. Intrathecal baclofen for treating spasticity in children with cerebral palsy. Cochrane Database Syst Rev. 2015;2015(11):CD004552.

  • 3

    Albright AL, Cervi A, Singletary J. Intrathecal baclofen for spasticity in cerebral palsy. JAMA. 1991;265(11):14181422.

  • 4

    Borowski A, Littleton AG, Borkhuu B, et al. Complications of intrathecal baclofen pump therapy in pediatric patients. J Pediatr Orthop. 2010;30(1):7681.

  • 5

    Desai VR, Raskin JS, Mohan A, et al. A standardized protocol to reduce pediatric baclofen pump infections: a quality improvement initiative. J Neurosurg Pediatr. 2018;21(4):395400.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Fjelstad AB, Hommelstad J, Sorteberg A. Infections related to intrathecal baclofen therapy in children and adults: frequency and risk factors. J Neurosurg Pediatr. 2009;4(5):487493.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Gooch JL, Oberg WA, Grams B, Ward LA, Walker ML. Complications of intrathecal baclofen pumps in children. Pediatr Neurosurg. 2003;39(1):16.

  • 8

    Haranhalli N, Anand D, Wisoff JH, et al. Intrathecal baclofen therapy: complication avoidance and management. Childs Nerv Syst. 2011;27(3):421427.

  • 9

    Imerci A, Rogers KJ, Pargas C, Sees JP, Miller F. Identification of complications in paediatric cerebral palsy treated with intrathecal baclofen pump: a descriptive analysis of 15 years at one institution. J Child Orthop. 2019;13(5):529535.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Lam SK, Mayer RR, Vedantam A, A Staggers K, Harris DA, Pan IW. Readmission and complications within 30 days after intrathecal baclofen pump placement. Dev Med Child Neurol. 2018;60(10):10381044.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Motta F, Antonello CE. Analysis of complications in 430 consecutive pediatric patients treated with intrathecal baclofen therapy: 14-year experience. J Neurosurg Pediatr. 2014;13(3):301306.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Tyack L, Copeland L, McCartney L, Waugh MC. Adverse events associated with paediatric intrathecal baclofen in Australia: 5 years of data collection. J Paediatr Child Health. 2021;57(2):258262.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Spader HS, Bollo RJ, Bowers CA, Riva-Cambrin J. Risk factors for baclofen pump infection in children: a multivariate analysis. J Neurosurg Pediatr. 2016;17(6):756762.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Kestle JR, Holubkov R, Douglas Cochrane D, et al. A new Hydrocephalus Clinical Research Network protocol to reduce cerebrospinal fluid shunt infection. J Neurosurg Pediatr. 2016;17(4):391396.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Kestle JR, Riva-Cambrin J, Wellons JC III, et al. A standardized protocol to reduce cerebrospinal fluid shunt infection: the Hydrocephalus Clinical Research Network Quality Improvement Initiative. J Neurosurg Pediatr. 2011;8(1):2229.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Forrest CB, Margolis PA, Bailey LC, et al. PEDSnet: a national pediatric learning health system. J Am Med Inform Assoc. 2014;21(4):602606.

  • 17

    Bailes AF, Kean J, Gross PH, et al. The Cerebral Palsy Research Network: building a learning health network for cerebral palsy. J Pediatr Rehabil Med. 2021;14(2):161171.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Langley G, Nolan K, Nolan T, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. 2nd ed. Joffrey-Bass; 2009.

  • 19

    Taylor MJ, McNicholas C, Nicolay C, Darzi A, Bell D, Reed JE. Systematic review of the application of the plan-do-study-act method to improve quality in healthcare. BMJ Qual Saf. 2014;23(4):290298.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Benneyan JC, Lloyd RC, Plsek PE. Statistical process control as a tool for research and healthcare improvement. Qual Saf Health Care. 2003;12(6):458464.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Provost L, Murray S. The Heath Care Data Guide,. Learning from Data for Improvement. Jossey-Bass; 2011.

  • 22

    Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360(5):491499.

  • 23

    Ryan SL, Sen A, Staggers K, Luerssen TG, Jea A. A standardized protocol to reduce pediatric spine surgery infection: a quality improvement initiative. J Neurosurg Pediatr. 2014;14(3):259265.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Schaffzin J, Simon K, Connelly B, Mangano F. Standardizing preoperative preparation to reduce surgical site infections among pediatric neurosurgical patients. J Neurosurg Pediatr. 2017;19(4):399406.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Tipper GA, Chiwera L, Lucas J. Reducing surgical site infection in pediatric scoliosis surgery: a multidisciplinary improvement program and prospective 4-year audit. Global Spine J. 2020;10(5):633639.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • Collapse
  • Expand
  • FIG. 1.

    Key driver diagram representing project to increase the number of protocol steps documented as completed in the operating room (OR) and to reduce infection rate. REDCap = Research Electronic Data Capture.

  • FIG. 2.

    Infection reduction protocol form. ABX = antibiotics; CHG = chlorhexidine gluconate; Gent = gentamicin; Vanc = vancomycin. © Cerebral Palsy Research Network, published with permission.

  • FIG. 3.

    Mean percentage of infection reduction protocol steps documented as completed over time. Figure is available in color online only.

  • FIG. 4.

    Percentage of pump infections within 30 days (y-axis) over time (x-axis). The preimplementation period included 2014–2017, and the postimplementation period was initiated in January 2018. No special cause variation was observed in the data. Figure is available in color online only.

  • 1

    What is cerebral palsy? Centers for Disease Control and Prevention. US Department of Health and Human Services. May 2, 2022. Accessed January 24, 2023. https://www.cdc.gov/ncbddd/cp/facts.html#:~:text=Cerebral%20palsy%20(CP)%20is%20a,problems%20with%20using%20the%20muscles

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Hasnat MJ, Rice JE. Intrathecal baclofen for treating spasticity in children with cerebral palsy. Cochrane Database Syst Rev. 2015;2015(11):CD004552.

  • 3

    Albright AL, Cervi A, Singletary J. Intrathecal baclofen for spasticity in cerebral palsy. JAMA. 1991;265(11):14181422.

  • 4

    Borowski A, Littleton AG, Borkhuu B, et al. Complications of intrathecal baclofen pump therapy in pediatric patients. J Pediatr Orthop. 2010;30(1):7681.

  • 5

    Desai VR, Raskin JS, Mohan A, et al. A standardized protocol to reduce pediatric baclofen pump infections: a quality improvement initiative. J Neurosurg Pediatr. 2018;21(4):395400.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Fjelstad AB, Hommelstad J, Sorteberg A. Infections related to intrathecal baclofen therapy in children and adults: frequency and risk factors. J Neurosurg Pediatr. 2009;4(5):487493.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Gooch JL, Oberg WA, Grams B, Ward LA, Walker ML. Complications of intrathecal baclofen pumps in children. Pediatr Neurosurg. 2003;39(1):16.

  • 8

    Haranhalli N, Anand D, Wisoff JH, et al. Intrathecal baclofen therapy: complication avoidance and management. Childs Nerv Syst. 2011;27(3):421427.

  • 9

    Imerci A, Rogers KJ, Pargas C, Sees JP, Miller F. Identification of complications in paediatric cerebral palsy treated with intrathecal baclofen pump: a descriptive analysis of 15 years at one institution. J Child Orthop. 2019;13(5):529535.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10

    Lam SK, Mayer RR, Vedantam A, A Staggers K, Harris DA, Pan IW. Readmission and complications within 30 days after intrathecal baclofen pump placement. Dev Med Child Neurol. 2018;60(10):10381044.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Motta F, Antonello CE. Analysis of complications in 430 consecutive pediatric patients treated with intrathecal baclofen therapy: 14-year experience. J Neurosurg Pediatr. 2014;13(3):301306.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Tyack L, Copeland L, McCartney L, Waugh MC. Adverse events associated with paediatric intrathecal baclofen in Australia: 5 years of data collection. J Paediatr Child Health. 2021;57(2):258262.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Spader HS, Bollo RJ, Bowers CA, Riva-Cambrin J. Risk factors for baclofen pump infection in children: a multivariate analysis. J Neurosurg Pediatr. 2016;17(6):756762.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Kestle JR, Holubkov R, Douglas Cochrane D, et al. A new Hydrocephalus Clinical Research Network protocol to reduce cerebrospinal fluid shunt infection. J Neurosurg Pediatr. 2016;17(4):391396.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Kestle JR, Riva-Cambrin J, Wellons JC III, et al. A standardized protocol to reduce cerebrospinal fluid shunt infection: the Hydrocephalus Clinical Research Network Quality Improvement Initiative. J Neurosurg Pediatr. 2011;8(1):2229.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Forrest CB, Margolis PA, Bailey LC, et al. PEDSnet: a national pediatric learning health system. J Am Med Inform Assoc. 2014;21(4):602606.

  • 17

    Bailes AF, Kean J, Gross PH, et al. The Cerebral Palsy Research Network: building a learning health network for cerebral palsy. J Pediatr Rehabil Med. 2021;14(2):161171.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Langley G, Nolan K, Nolan T, et al. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. 2nd ed. Joffrey-Bass; 2009.

  • 19

    Taylor MJ, McNicholas C, Nicolay C, Darzi A, Bell D, Reed JE. Systematic review of the application of the plan-do-study-act method to improve quality in healthcare. BMJ Qual Saf. 2014;23(4):290298.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Benneyan JC, Lloyd RC, Plsek PE. Statistical process control as a tool for research and healthcare improvement. Qual Saf Health Care. 2003;12(6):458464.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Provost L, Murray S. The Heath Care Data Guide,. Learning from Data for Improvement. Jossey-Bass; 2011.

  • 22

    Haynes AB, Weiser TG, Berry WR, et al. A surgical safety checklist to reduce morbidity and mortality in a global population. N Engl J Med. 2009;360(5):491499.

  • 23

    Ryan SL, Sen A, Staggers K, Luerssen TG, Jea A. A standardized protocol to reduce pediatric spine surgery infection: a quality improvement initiative. J Neurosurg Pediatr. 2014;14(3):259265.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Schaffzin J, Simon K, Connelly B, Mangano F. Standardizing preoperative preparation to reduce surgical site infections among pediatric neurosurgical patients. J Neurosurg Pediatr. 2017;19(4):399406.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Tipper GA, Chiwera L, Lucas J. Reducing surgical site infection in pediatric scoliosis surgery: a multidisciplinary improvement program and prospective 4-year audit. Global Spine J. 2020;10(5):633639.

    • PubMed
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 1358 951 45
Full Text Views 474 330 10
PDF Downloads 415 331 11
EPUB Downloads 0 0 0