Volume 30, Supplement - Infectious Diseases and Carceral Health
SUPPLEMENT ISSUE
Outbreaks and Investigations
Outbreak of Invasive Serratia marcescens among Persons Incarcerated in a State Prison, California, USA, March 2020–December 2022
Abstract
Serratia marcescens is an environmental gram-negative bacterium that causes invasive disease in rare cases. During 2020–2022, an outbreak of 21 invasive Serratia infections occurred in a prison in California, USA. Most (95%) patients had a history of recent injection drug use (IDU). We performed whole-genome sequencing and found isolates from 8 patients and 2 pieces of IDU equipment were closely related. We also identified social interactions among patients. We recovered S. marcescens from multiple environmental samples throughout the prison, including personal containers storing Cell Block 64 (CB64), a quaternary ammonium disinfectant solution. CB64 preparation and storage conditions were suboptimal for S. marcescens disinfection. The outbreak was likely caused by contaminated CB64 and propagated by shared IDU equipment and social connections. Ensuring appropriate preparation, storage, and availability of disinfectants and enacting interventions to counteract disease spread through IDU can reduce risks for invasive Serratia infections in California prisons.
Serratia marcescens, a gram-negative environmental bacterium (1,2), is an opportunistic pathogen that in rare cases causes invasive diseases, including bacteremia and endocarditis (1,3–7). Reported outbreaks have been linked to contaminated environmental sources, such as water, soap, intravenous fluids, and compounded drugs (8–16) in nosocomial settings (17–19). Invasive S. marcescens infections have occurred among persons who inject drugs (5,6,20–22). Given the high prevalence of injection drug use (IDU) in prisons and lack of access to sterile needles (23–25), risks for transmission of bloodborne pathogens are higher than among the general population (25). Cell Block 64 (CB64) solution, produced by California Prison Industry (CALPIA, https://www.calpia.ca.gov), is a quaternary ammonium concentrate (https://catalog.calpia.ca.gov/custom/assets/Files/view-current-sds-information-16.pdf) used as the primary disinfectant in prisons in California, USA. However, S. marcescens can survive in improperly prepared disinfection solutions, including quaternary ammonium disinfectants (18,19,26).
We describe a multiyear outbreak of invasive S. marcescens infections driven by widespread environmental contamination, improperly prepared and maintained disinfection solution, IDU, and social connections at a California state prison. Prison A is a maximum-security state prison housing ≈3,000 male incarcerated persons. In October 2020, the primary hospital affiliated with prison A notified the California Correctional Health Care System (CCHCS) that multiple incarcerated persons had been admitted with invasive S. marcescens infections. CCHCS, Monterey County Public Health Laboratory (MCPHL), and California Department of Public Health (CDPH) began a multidisciplinary investigation to identify additional cases, determine risk factors for infection, and provide recommendations for mitigation and prevention. This project was determined to be nonresearch by the Centers for Disease Control and Prevention because it involved public health surveillance.
Epidemiologic Investigation
We defined a case-patient as a person diagnosed with an invasive S. marcescens infection who resided at prison A for ≥1 month before symptom onset during January 1, 2020–December 31, 2022. We defined infections as invasive if occurring at normally sterile body sites or in a case-patient manifesting critical illness with severe soft tissue infection. We reviewed patient hospitalization and prison medical records, including social histories, for IDU and other risk-elevating behaviors. We interviewed patients using a standardized questionnaire that included questions about cell cleaning practices, IDU, and other risk factors.
Environmental Investigation
Prison A public health and infection control, CCHCS public health, CDPH, and MCPHL staff evaluated the water system and cleaning practices and procedures at prison A. In 2020 and 2021, MCPHL tested water from different sources at prison A, including holding tanks and wells. MCPHL also tested sinks and communal showers, faucets in patients’ cells, personal items, hand-rinsate from a cellmate, 2 syringes used for injecting drugs, objects used for mixing, storing, or applying disinfectant, dilution machines, reused containers, and commercial bottles.
Laboratory Investigation
MCPHL streaked swabs onto Serratia CHROMagar (https://www.chromagar.com) MacConkey and blood agar plates and incubated them in brain heart infusion broth for up to 5 days. Needles and syringes were placed directly into brain heart infusion broth. Cultures with growth were subcultured on CHROMagar plates. Liquids, including water, disinfectant cleaning solutions, and rinsates, were filtered onto 47 mm 0.45 μm–pore sized mixed cellulose ester membranes and placed onto CHROMagar plates. MCPHL forwarded S. marcescens isolates to CDPH Center for Laboratory Science Microbial Diseases Laboratory for whole genome sequencing (WGS) using the validated in-house protocol with Illumina MiSeq (https://www.illumina.com) (Appendix) (27).
Epidemiologic Investigation
As of December 2022, we had identified 21 cases: 17 identified during March 2020–August 2021 and 4 during April–October 2022 (Figure 1). All 21 case-patients were hospitalized and recovered; however, 1 patient later died of a cause unrelated to S. marcescens. Median patient age was 44 years (range 22–66 years). We grouped patients by race/ethnicity as non-Hispanic White (9 [43%]), non-Hispanic Black (2 [10%]), Hispanic (8 [38%]), or other (2 [10%]). Diagnoses were not mutually exclusive and included bacteremia in 11 (52%) patients; endocarditis in 2 (10%); epidural abscess in 9 (43%); osteomyelitis in 6 (29%); pseudoaneurysm in 1 (5%); and soft tissue infections in 4 (19%), including 2 (10%) with muscle abscess (Table 1). Of the nonbacteremic patients, 2 had polymicrobial cultures, including viridans streptococci (1), Staphylococcus aureus (2), and Raoutella panticola (1).
Twenty (95%) patients had a history of IDU <6 months before infection and one >6 months before infection. Of patients with recent IDU, 18/20 (86%) had injected heroin, 12 (57%) suboxone, and 8 (38%) methamphetamines. Among patients who had urine toxicology performed at admission, 4/9 were positive for opiates. Nine patients reported consuming >1 drug; 5 patients used 2 and 4 patients used 3 drugs. Of patients interviewed, 5/16 (31%) used CB64 to clean IDU equipment. Of PWID patients, 9/21 (43%) were enrolled in the prison A substance use disorder treatment (SUDT) program before S. marcescens infection occurred.
Although some patients resided throughout the 4 physically separated yards at the facility, 11 (52%) were housed in yard 1; 2 patients in other yards at time of illness onset had previously been housed in yard 1. Interviews identified social connections among >9 patients. We used WGS to identify the predominant S. marcescens outbreak strain as the cause of infection in 6 (66%) patients and a different strain in 1 patient; we had no isolates available for 4 patients (Figures 2, 3). Among patients who revealed social connections, 6 shared needles, 4 shared cells, 3 had attended the urgent care clinic at the same time, and 2 might have shared tattoo needles (Figure 2). Patient AC, diagnosed with an invasive S. marcescens infection at a different prison in 2019 and later transferred to prison A, was found to have multiple social connections with patients identified in 2020 and beyond but we did not include him in the outbreak cohort (Figure 2).
Environmental Investigation
Inspection of the potable water system at prison A did not identify any deficiencies or areas of concern, and we found large-volume water samples negative for Serratia. Each housing unit has a machine for diluting the CB64 solution (Figure 4). Machines in multiple units had exposed tubing touching the machine surface, maintenance schedules were not documented, and dilution of CB64 occurred in large containers outside the dilution machines. Prison A allowed incarcerated persons to keep CB64 in their cells after the COVID-19 pandemic started. Some incarcerated persons described using their own repurposed containers (e.g., shampoo bottles) to scoop diluted CB64 from the large containers.
Laboratory Investigation
Eleven patients had isolates available for WGS; 8 (73%) patients, including 3 identified in 2022, had isolates that differed from one another by 0–19 single-nucleotide polymorphisms (SNPs) on WGS. Those isolates clustered within the predominant strain group (Figure 3). Of 152 environmental samples collected and analyzed, 27 (18%) were positive for S. marcescens, including a needle and syringe combination (sample I) and a reused nasal spray bottle (sample C) storing methamphetamines from patient D (Table 2). Both specimens matched the predominant outbreak strain (Figure 3). The S. marcescens isolate from patient F grouped within 11–17 SNPs with isolates from a coffee cup (sample B) found in patient D’s cell, hand-rinsate (sample H) from patient D’s cellmate, and a doorway swab (sample Y) from a cell occupied at different times by patients AA and AC (Table 2; Figure 3). All other isolates differed from the predominant strain by thousands of SNPs (Figure 3). We sequenced multiple isolates from some samples. Samples from all unopened bottles of CB64 tested negative for S. marcescens.
During March 2020–December 2022, a total of 21 persons incarcerated at prison A required hospitalization for invasive Serratia infections. Factors contributing to this outbreak included widespread environmental contamination with Serratia, including in CB64, the sole disinfectant used within the prison, and complex social networks that involved IDU.
Of note, 5 environmental samples that tested positive for Serratia were associated with diluted CB64. CB64 is used throughout California state prisons as a disinfectant because it is less caustic than other disinfectants (e.g., bleach). Quaternary ammonium compounds like CB64 have previously been linked to outbreaks (18,19,26). Prison A had documented nonadherence to CB64 manufacturer dilution and storage protocols. In addition, incarcerated persons stored diluted CB64 in cells after the COVID-19 pandemic began, a change in procedure occurring at approximately the same time as initial cases. Repeatedly finding Serratia in CB64 indicates that improper use and storage of the disinfectant likely contributed to the spread.
The invasive nature of the Serratia infections, including manifestations such as bacteremia and severe soft-tissue infection, suggests introduction of the bacteria directly into the bloodstream or soft tissues, highlighting the role of IDU in the prison outbreak. The predominant outbreak strain of Serratia was recovered from a needle obtained from 1 patient. In prisons, there is no access to new needles; some patients reported sharing needles, and most reported reusing needles multiple times themselves. Some patients reported using CB64 to clean their needles.
In August 2021, prison A implemented mitigation measures, including extensive staff training, instituting maintenance logs, recalibrating dilution machines, ensuring regular changing of tubing in dilution devices, and providing dedicated bottles of CB64 for incarcerated persons to check out and return within 24 hours for in-cell cleaning. Additional education on IDU risks and SUDT (begun in 2020) were also provided to incarcerated persons. No new cases were identified until spring 2022, at which time lapses in staff and resident education on use, maintenance, and storage of CB64 solution and dilution devices were recognized.
WGS results for 3 patient isolates identified at the prison in 2022 were closely related to 2021 patient isolates, indicating that the predominant outbreak strain of S. marcescens persisted >1 year. Given the diversity of S. marcescens strains in the environment, the predominance of a single strain suggests the likely existence of a persistent, but unknown, nidus of the outbreak strain. A single contaminated drug or CB64 source is unlikely to account for the persistence. An incarcerated person colonized with this strain or an unrecognized fomite in the environment are possible sources. Although S. marcescens is not a normal part of human flora, colonization of skin and gut has been documented (10,27). In addition, the hand-rinsate from a patient’s cellmate yielded S. marcescens, indicating the potential for persistence on skin. After identification of additional cases in 2022, intervention included reeducating staff and incarcerated persons on proper use of CB64, including performing dilution within dilution devices only, and education on risks for S. marcescens infection through IDU equipment. No further cases had been documented as of July 2023, 8 months after the last identified case. Additional education has been provided to institutions throughout the state (Appendix).
One limitation of this study is that, given drug use is prohibited in prison, patients might have provided incomplete information regarding drug preparation and sharing, and therefore some common sources of drugs or drug equipment may not have been identified. In addition, only 2 needles or syringes were available for testing. A comprehensive environmental sampling survey of the entire prison population and structure was unfeasible, so we focused testing on areas where cases were identified. Additional sources of environmental contamination, including water sources such as cell toilet water and shower and sink drains and traps, where biofilm may have formed, were unable to be tested. A limited number of patient isolates from 2021 and 2022 were available for WGS; testing of all isolates might have further clarified patient connections. Patients might have been infected with >1 S. marcescens strain. Most environmental isolates positive for S. marcescens did not match patient strains, and so direct correlation between environmental contamination and patient illness was not possible. Finally, our investigation focused on invasive infections and excluded milder illness.
Beginning in January 2020, screening and referral for SUDT became available in California prisons to all newly incarcerated persons, those transitioning into the community, and patients with IDU-related complications (28). As of January 2022, >64,600 incarcerated persons had been screened for SUD and medication-assisted treatment provided to >22,500 patients, leading to a significantly decline in overdoses and infectious disease complications since the program started (29).
After this outbreak, queries have identified additional cases of invasive S. marcescens infections in other California prisons. Similar concerns related to disinfection, including improper storage, device calibration, and usage, and IDU practices have been reported. Environmental mitigation through extensive cleaning and strict adherence to disinfectant guidelines might not eliminate all environmental sources of Serratia but might decrease the environmental microbial burden, thereby decreasing potential exposures to S. marcescens and other pathogens. IDU among incarcerated persons should be addressed through promotion of harm reduction practices and education, including access to appropriate disinfection supplies and sterile needles, and referral to SUDT programs.
Dr. Kamali is a medical epidemiologist and infectious disease physician with California Correctional Health Care Services and California Department of Public Health whose interests include outbreak investigation.
Acknowledgments
We thank Allen Radner, Andrew Massengill, Darlene Brieno, Lucy Tompkins, Stanley Deresinski, Jonathon Yoder, Matthew Arduino, Mia Mattioli, and Jon Rosenberg. We also thank the incarcerated persons, patients, and staff of the California Department of Corrections and Rehabilitation (prison A).
Laboratory investigation was supported the Centers for Disease Control and Prevention through a grant on the epidemiology and laboratory capacity for infectious diseases from (grant no. 5NU50CK000539) and by the California General Fund.
Author contributions: concept/design: A.Ka., N.O., H.B., A.Y., and A.Ki.; data acquisition: A.Ka., D.F., H.D., N.O., R.M., R.L., J.C., A.Y., and A.Ki.; data analysis: A.Ka., D.F., C.S., R.L., J.H., J.C., A.Y., and A.Ki.; draft: A.Ka., C.S., A.Y., and A.Ki.; revision: A.Ka., D.F., H.D., N.O., R.M., C.S., R.L., J.H., M.S., H.B., A.Y., and A.Ki.
References
- Fisher RG. Serratia. In: Feigin RD, Cherry JD, Demmler-Harrison GJ, Kaplan SL, eds. Feigin and Cherry’s textbook of pediatric infectious diseases (6th edition). Philadelphia: W.B. Saunders, 2009. pp. 1563–7.
- Donnenberg MS. Enterobacteriaceae. Bennett JE, Dolin R, and Blaser MJ, eds. Mandell, Douglas, and Bennett’s principles and practice of infectious diseases, 8th edition. Philadelphia: W.B. Saunders, 2015. p. 2503–17.e5.
- Mahlen SD. Serratia infections: from military experiments to current practice. Clin Microbiol Rev. 2011;24:755–91. DOIPubMedGoogle Scholar
- Rana A, Rabbani NUA, Wood S, McCorkle C, Gilkerson C. A complicated case of vertebral osteomyelitis by Serratia marcescens. Cureus. 2020;12:
e9002 . DOIPubMedGoogle Scholar - Sanchez KT, Johnson LB, Szpunar S, Saravolatz LD. A case of mixed Serratia marcescens and Streptococcus mitis endocarditis and review of the literature. Infect Dis Clin Pract. 2012;20:245–7. DOIGoogle Scholar
- Schecter MC, Spicer JO, Aldrete SDM, Kraft CS. Serratia marcescens infectious endocarditis injection drug use, left-sided heart disease, and poor outcomes. Infect Dis Clin Pract. 2018;26:216–9. DOIGoogle Scholar
- Hadid H, Usman M, Thapa S. Severe osteomyelitis and septic arthritis due to Serratia marcescens in an immunocompetent patient. Case Rep Infect Dis. 2015;2015:
347652 . DOIPubMedGoogle Scholar - Horcajada JP, Martínez JA, Alcón A, Marco F, De Lazzari E, de Matos A, et al. Acquisition of multidrug-resistant Serratia marcescens by critically ill patients who consumed tap water during receipt of oral medication. Infect Control Hosp Epidemiol. 2006;27:774–7. DOIPubMedGoogle Scholar
- Civen R, Vugia DJ, Alexander R, Brunner W, Taylor S, Parris N, et al. Outbreak of Serratia marcescens infections following injection of betamethasone compounded at a community pharmacy. Clin Infect Dis. 2006;43:831–7. DOIPubMedGoogle Scholar
- Cristina ML, Sartini M, Spagnolo AM. Serratia marcescens infections in neonatal intensive care units (NICUs). Int J Environ Res Public Health. 2019;16:610. DOIPubMedGoogle Scholar
- Caggiano G, Triggiano F, Diella G, Apollonio F, Lopuzzo M, Mosca A, et al. A possible outbreak by Serratia marcescens: genetic relatedness between clinical and environmental strains. Int J Environ Res Public Health. 2021;18:9814. DOIPubMedGoogle Scholar
- Archibald LK, Corl A, Shah B, Schulte M, Arduino MJ, Aguero S, et al. Serratia marcescens outbreak associated with extrinsic contamination of 1% chlorxylenol soap. Infect Control Hosp Epidemiol. 1997;18:704–9. DOIPubMedGoogle Scholar
- Henry B, Plante-Jenkins C, Ostrowska K. An outbreak of Serratia marcescens associated with the anesthetic agent propofol. Am J Infect Control. 2001;29:312–5. DOIPubMedGoogle Scholar
- Ostrowsky BE, Whitener C, Bredenberg HK, Carson LA, Holt S, Hutwagner L, et al. Serratia marcescens bacteremia traced to an infused narcotic. N Engl J Med. 2002;346:1529–37. DOIPubMedGoogle Scholar
- Boyce JM, Havill NL. In-use contamination of a hospital-grade disinfectant. Am J Infect Control. 2022;50:1296–301. DOIPubMedGoogle Scholar
- Weber DJ, Rutala WA, Sickbert-Bennett EE. Outbreaks associated with contaminated antiseptics and disinfectants. Antimicrob Agents Chemother. 2007;51:4217–24. DOIPubMedGoogle Scholar
- de Boer MG, Brunsveld-Reinders AH, Salomons EM, Dijkshoorn L, Bernards AT, van den Berg PCM, et al. Multifactorial origin of high incidence of Serratia marcescens in a cardio-thoracic ICU: analysis of risk factors and epidemiological characteristics. J Infect. 2008;56:446–53. DOIPubMedGoogle Scholar
- Rohit A, Suresh Kumar D, Dhinakaran I, Joy J, Vijay Kumar D, Kumar Ballamoole K, et al. Whole-genome-based analysis reveals multiclone Serratia marcescens outbreaks in a non-Neonatal Intensive Care Unit setting in a tertiary care hospital in India. J Med Microbiol. 2019;68:616–21. DOIPubMedGoogle Scholar
- de Vries JJ, Baas WH, van der Ploeg K, Heesink A, Degener JE, Arends JP. Outbreak of Serratia marcescens colonization and infection traced to a healthcare worker with long-term carriage on the hands. Infect Control Hosp Epidemiol. 2006;27:1153–8. DOIPubMedGoogle Scholar
- Peterson TC, Pearson C, Zekaj M, Hudson I, Fakhouri G, Vaidya R. Septic arthritis in intravenous drug abusers: a historical comparison of habits and pathogens. J Emerg Med. 2014;47:723–8. DOIPubMedGoogle Scholar
- Phadke VK, Jacob JT. Marvelous but Morbid: Infective endocarditis due to Serratia marcescens. [Baltim Md]. Infect Dis Clin Pract (Baltim Md). 2016;24:143–50. DOIPubMedGoogle Scholar
- Cooper R, Mills J. Serratia endocarditis. A follow-up report. Arch Intern Med. 1980;140:199–202. DOIPubMedGoogle Scholar
- Bick JA. Infection control in jails and prisons. Clin Infect Dis. 2007;45:1047–55. DOIPubMedGoogle Scholar
- Davis DM, Bello JK, Rottnek F. Care of incarcerated patients. Am Fam Physician. 2018;98:577–83.PubMedGoogle Scholar
- Hammett TM. HIV/AIDS and other infectious diseases among correctional inmates: transmission, burden, and an appropriate response. Am J Public Health. 2006;96:974–8. DOIPubMedGoogle Scholar
- Nakashima AK, Highsmith AK, Martone WJ. Survival of Serratia marcescens in benzalkonium chloride and in multiple-dose medication vials: relationship to epidemic septic arthritis. J Clin Microbiol. 1987;25:1019–21. DOIPubMedGoogle Scholar
- Kozyreva VK, Truong CL, Greninger AL, Crandall J, Mukhopadhyay R, Chaturvedi V. Validation and implementation of Clinical Laboratory Improvements Act—compliant whole-genome sequencing in the public health microbiology laboratory. J Clin Microbiol. 2017;55:2502–20. DOIPubMedGoogle Scholar
- Jones SR, Amon M, Falvey C, Patrick K. Serratia marcescens colonising the gut. Lancet. 1978;1:1105. DOIPubMedGoogle Scholar
- California Department of Corrections and Rehabilitation, California Correctional Health Care Services. Transforming substance use disorder treatment in California’s prison system. Impacts of the Integrated Substance Use Disorder Treatment Program 2019–2021, April 2022 [cited 2023 May 1]. https://cchcs.ca.gov/wp-content/uploads/sites/60/ISUDT/Impacts-ISUDT-Program2019-22.pdf
Figures
Tables
Cite This ArticleTable of Contents – Volume 30, Supplement—March 2024
EID Search Options |
---|
|
|
|
Please use the form below to submit correspondence to the authors or contact them at the following address:
Amanda Kamali, California Correctional Health Care Services, 8260 Longleaf Dr, Elk Grove, CA 95758, USA
Top