American Journal of Infection Control 44 (2016) 269-72
Contents lists available at ScienceDirect
American Journal of Infection Control
American Journal of Infection Control
journal homepage: www.ajicjournal.org
Major article
Nosocomial transmission of Ebola virus disease on pediatric and maternity wards: Bombali and Tonkolili, Sierra Leone, 2014 Angela C. Dunn MD a, b, *, Tiffany A. Walker MD a, b, John Redd MD b, David Sugerman MD b, Jevon McFadden MD b, Tushar Singh MD a, b, Joseph Jasperse MPH c, Brima Osaio Kamara MD d, Tom Sesay MD e, James McAuley MD b, Peter H. Kilmarx MD b a
Epidemic Intelligence Service, Centers for Disease Control and Prevention, Atlanta, GA CDC Sierra Leone Ebola Response Team, Freetown, Sierra Leone Concern Worldwide, Tonkolili District, Sierra Leone d Sierra Leone Ministry of Health and Sanitation, Tonkolili District, Sierra Leone e Sierra Leone Ministry of Health and Sanitation, Bombali District, Sierra Leone b c
Key Words: West Africa Hospital-acquired infection Infection control Epidemiology Outbreak
Background: In the largest Ebola virus disease (EVD) outbreak in history, nosocomial transmission of EVD increased spread of the disease. We report on 2 instances in Sierra Leone where patients unknowingly infected with EVD were admitted to a general hospital ward (1 pediatric ward and 1 maternity ward), exposing health care workers, caregivers, and other patients to EVD. Both patients died on the general wards, and were later confirmed as being infected with EVD. We initiated contact tracing and assessed risk factors for secondary infections to guide containment recommendations. Methods: We reviewed medical records to establish the index patients’ symptom onset. Health care workers, patients, and caregivers were interviewed to determine exposures and personal protective equipment (PPE) use. Contacts were monitored daily for EVD symptoms. Those who experienced EVD symptoms were isolated and tested. Results: Eighty-two contacts were identified: 64 health care workers, 7 caregivers, 4 patients, 4 newborns, and 3 children of patients. Seven contacts became symptomatic and tested positive for EVD: 2 health care workers (1 nurse and 1 hospital cleaner), 2 caregivers, 2 newborns, and 1 patient. The infected nurse placed an intravenous catheter in the pediatric index patient with only short gloves PPE and the hospital cleaner cleaned the operating room of the maternity ward index patient wearing short gloves PPE. The maternity ward index patient’s caregiver and newborn were exposed to her body fluids. The infected patient and her newborn shared the ward and latrine with the maternity ward index patient. Hospital staff members did not use adequate PPE. Caregivers were not offered PPE. Conclusions: Delayed recognition of EVD and inadequate PPE likely led to exposures and secondary infections. Earlier recognition of EVD and adequate PPE might have reduced direct contact with body fluids. Limiting nonhealth-care worker contact, improving access to PPE, and enhancing screening methods for pregnant women, children, and inpatients may help decrease EVD transmission in general health care settings. Published by Elsevier Inc. on behalf of the Association for Professionals in Infection Control and Epidemiology, Inc.
* Address correspondence to Angela C. Dunn, MD, Bureau of Epidemiology, Utah Department of Health, PO Box 142104, Salt Lake City, UT 84114-2104. E-mail address: [email protected] (A.C. Dunn). ACD and TAW contributed equally to this work. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the Centers for Disease Control and Prevention or the institutions with which the authors are affiliated. Conflicts of interest: None.
During the largest Ebola (EVD) outbreak in history, health care workers became infected with EVD while delivering patient care. As of October 31, 2014, 199 of 3,854 (5.2%) laboratory-confirmed EVD cases reported from Sierra Leone were in health care workers.1 We report on 2 patients who were admitted to and died on general hospital wards in October 2014 and were confirmed postmortem as being EVD positive via oral swab by the Centers for Disease Control and Prevention (CDC) Sierra Leone EVD laboratory,2 potentially exposing health care workers to EVD. In Bombali
0196-6553/$00.00 - Published by Elsevier Inc. on behalf of the Association for Professionals in Infection Control and Epidemiology, Inc. http://dx.doi.org/10.1016/j.ajic.2015.09.016
270
A.C. Dunn et al. / American Journal of Infection Control 44 (2016) 269-72
District, a child was admitted to a pediatric ward for presumptive malaria and in Tonkolili District, a pregnant woman was admitted to a maternity ward for prolonged labor. Both patients were screened for EVD on admission; however, neither patient was reassessed for symptoms of EVD after admission. Both patients later died from undiagnosed EVD on the general wards, confirmed by postmortem testing. Both hospitals used the Sierra Leone Ministry of Health and Sanitation EVD case definition to triage patients on admission. The Sierra Leone EVD case definition defines a suspected EVD patient as any person with fever (>38 C) plus 3 or more of the following symptoms: vomiting, headache, nausea, diarrhea, difficulty breathing, fatigue, abdominal pain, loss of appetite, muscle or joint pain, unexplained bleeding, difficulty swallowing, or hiccups. Or, a suspected EVD patient is any ill person with an epidemiologic link to a suspected or confirmed EVD case.3 The CDC assisted the District Health Management Teams in investigating these health care-related EVD virus exposures. Those who had direct contact with EVD-infected patients were monitored for signs and symptoms of EVD over the 21-day incubation period. EVD was diagnosed at the CDC-Sierra Leone EVD laboratory by testing serum using reverse transcriptionepolymerase chain reaction (RT-PCR).2 Occupational exposures of those who became infected with EVD in this cohort included cleaning body fluids and placing intravenous catheters. It is essential that health care workers adhere to standard precautions3,4 when interacting with all patients. Both health care workers and caregivers need access to recommended personal protective equipment (PPE) and training on proper use. During an EVD outbreak, the following might help decrease EVD transmission in general health care settings: adhering to World Health Organization-recommended PPE guidelines, implementing effective EVD screening methods, and reassessing admitted patients for EVD. BOMBALI DISTRICT On October 8, 2014, a 29-month-old child presented with vomiting and fatigue to the pediatric outpatient center of the district’s government hospital. The patient did not meet the EVD screening case definition and therefore was admitted to the general pediatric ward, where he was treated for malaria. The patient developed diarrhea later that day. On October 10, he developed fever, rash, red eyes, and melena. The patient died October 13 on the pediatric ward. Given the patient’s symptoms, the attending pediatrician requested a serum specimen be drawn for EVD RT-PCR at the time of death and the results confirmed EVD. A total of 39 health care workers had contact with the source patient, including 32 nurses, 3 cleaners, 2 laboratory technicians, 1 mortuary worker, and 1 porter. CDC performed a review of staffing records and conducted in-person interviews with 28 (72%) of the health care worker contacts to determine the dates and circumstances of their exposure to the index patient. Thirty-six (92%) health care worker contacts lacked access to proper PPE; however, all 3 cleaners had access to and wore World Health Organizationrecommended PPE.3 There were 11 (31%) exposed nurses who were unavailable for interview due to author’s time constraints. A summary of exposures is presented in Table 1. At the direction of the hospital’s medical superintendent, the 36 health care workers with confirmed direct contact were placed on home quarantine for 21 days. The 3 cleaners who wore adequate PPE were excluded from home quarantine. A line list of exposed health care workers was submitted to the District Health Management Office. Contact tracers were assigned to each health care worker for daily direct monitoring of signs and symptoms of EVD. Those who developed symptoms were isolated in a designated room at the district’s government hospital. Their blood
specimens were sent to the CDC-Sierra Leone EVD Laboratory for testing. Of the 36 contacts, 1 health care worker developed symptoms 7 days after the occupational exposure and tested positive for EVD 6 days later, resulting in a health care worker attack rate of 2.8%. The infected individual was a nurse whose only reported exposure was placing an intravenous catheter in the index patient while wearing short gloves only. Exposure and outcome data for this nurse are presented in Table 2. TONKOLILI DISTRICT A 25-year-old full term pregnant woman presented with labor pains to her community maternity clinic on October 7, 2014. On October 9, she was transferred to the maternity ward of a general hospital due to prolonged labor where she was observed for 5 days. On October 14, the patient gave birth to a live newborn via an uncomplicated cesarean section for prolonged labor. Two days later, on October 16, she reported feeling febrile and had 1 episode of vomiting. The following day, October 17, the patient developed red eyes and had excessive bleeding from her surgical site and vagina. The physician noted in the patient’s chart that he would discuss isolation with the care team; meanwhile, the patient remained on the general maternity ward. The patient passed away on the general maternity ward October 18. Consistent with Sierra Leone national protocol,5 a postmortem buccal swab for EVD RT-PCR was performed and the results confirmed EVD. At the direction of the hospital’s medical superintendent and the district medical officer, all hospital staff, patients, and caregivers who came into direct contact with the index patient were quarantined in the maternity ward for 21 days. The medical superintendent assigned 1 staff member to screen the quarantined staff twice daily, including monitoring symptoms and checking body temperatures using an infrared thermometer. Screening data were recorded in a logbook that was reviewed in the mornings and evenings by the medical superintendent. The medical superintendent took the temperatures and assessed symptoms of the quarantined patients and their caregivers twice daily. Those persons who developed symptoms consistent with EVD were isolated in a predetermined, individual hospital room. Their blood specimens were sent to the CDC-Sierra Leone EVD Laboratory for testing. In addition, the district medical officer placed the contacts from the community maternity clinicd3 nursesdon home quarantine and active surveillance with daily assessments by a contact tracer for 21 days. A total of 46 people reportedly came into contact with the index patient from the time of her presentation to the community maternity clinic on October 7, including 25 hospital staff: 18 nurses, 5 hospital cleaners, 1 physician, and 1 laboratory technician. In addition, 7 family caregivers, 4 adult patients, 4 newborns, 3 children, and 3 community maternity clinic nurses were exposed. We reviewed medical records of the index patient to establish the timeline of events, and conducted in-person interviews with 37 contacts (80%) to determine their exposure history. Maternity ward nurses and hospital cleaning staff lacked access to recommended PPE and caregivers were not offered PPE. Out of the 8 exposed persons (10%) who were unable to be interviewed, 6 were children, 1 was the caregiver of the index patient who was at an EVD treatment unit with the index patient’s newborn, and 2 were hospital staff who were unable to be located. Maternity ward health care workers provided exposure history for these 8 contacts. A summary of exposures is presented in Table 1. Of the 46 contacts, 6 tested positive for EVD within 21 days of exposure to the index patient, resulting in a health care facility attack rate of 13.0%. Those who became infected were the newborn and caregiver of the index patient, 1 maternity ward patient and her newborn and caregiver,
A.C. Dunn et al. / American Journal of Infection Control 44 (2016) 269-72
271
Table 1 Exposures to patients with Ebola virus disease (EVD) and personal protective equipment (PPE) worn by contacts while exposed to patients with EVD, Bombali and Tonkolili Districts, Sierra Leone, October 2014
Exposure
Contacts with exposure (N ¼ 82)*
PPE worn while exposed
Shared ward/latrine Newborn of index EVD patient Newborn of confirmed EVD maternity ward patient Contact with newborn of an EVD patient Took vital signs Cleaned linens Vaginal exam Performed/assisted in cesarean section Cleaned body fluids Cleaned body fluids Cleaned surfaces: floor, walls, bed Cleaned surgical instruments Moved patient Placed urinary catheter Placed intravenous line Gave intravenous medications Blood draw Discontinued intravenous line Gave intramuscular medications Changed surgical site dressing General touching patient General touching patient Death pronouncement Unknown contact No known direct contact
None None None None Short gloves Short gloves Short gloves Gown; short gloves (three pairs); mask; goggles; shoe covers Short gloves None Gown; apron; short gloves (2 pairs); mask Short gloves Short gloves Short gloves; gown Short gloves Short gloves; gown Gown; apron; short gloves (2 pairs); mask Short gloves Short gloves Short gloves; gown Short gloves None Short gloves Unknown None
15 1 1 4 6 2 8 3 4 1 3 1 4 1 9 15 4 1 1 3 5 4 1 13 4
(18) (1) (1) (5) (7) (2) (10) (4) (5) (1) (4) (1) (5) (1) (11) (18) (5) (1) (1) (4) (6) (4) (1) (16) (5)
Contacts with exposure who developed EVD* 3 1 1 2 1
1 1
1 1
2
(20) (100) (100) (50) 0 (50) 0 0 (25) (100) 0 0 (25) 0 (11) 0 0 0 0 0 0 (50) 0 0 0
NOTE. Values are presented as n (%). *Some contacts had multiple exposures.
Table 2 Exposures and outcomes of contacts who developed Ebola virus disease, Bombali and Tonkolili Districts, Sierra Leone, October 2014 Contact ID
District
Relationship with index patient
1 2
Bombali Tonkolili
Nurse Newborn
3 4
Tonkolili Tonkolili
5*
Tonkolili
Caregiver Shared ward and latrine Shared ward
6*
Tonkolili
7
Tonkolili
Shared ward and latrine Hospital cleaner
Exposure(s)
Date(s) of exposure(s) to index patient
PPE worn while exposed
Date of symptom onset
Outcome
Placed intravenous catheter Intrauterine, cesarean section, breast milk, body contact Cleaned body fluids Body contact
October 10 Antenatal, October 14-18 October 9-18 October 9-18
Short gloves None
October 17 October 17
Death on October 27 Death on October 24
None None
November 1 October 22
Death on November 10 Death on October 28
Newborn of contact 4
Antenatal; October 16-18 October 9-18
None
October 23
Death on unknown date
None
Unknown
Death on unknown date
October 14
Short gloves
October 22
Death on unknown date
Body contact, caregiver for contact 4 Cleaned body fluids; moved patient; cleaned linens
*Contacts 5 and 6 might have contracted Ebola virus disease from contact 4 rather than the index patient, and thus might be secondary cases.
and 1 hospital cleaner. The newborn of the index patient was likely infected via mother-to-child transmission. The newborn of the infected maternity ward patient might have been infected via his/her mother’s breast milk. A summary of exposures and outcomes of those who contracted EVD is presented in Table 2. DISCUSSION Unidentified EVD patients in non-EVD health care facilities present a significant health risk to health care workers, caregivers, and other patients.6 Effective infection prevention and control measures to reduce the risk of nosocomial transmission to health care workers and patients include improved disease screening, isolation of suspect case patients, facility engineering controls, and adherence to recommended PPE.4 In the outbreaks we describe, screening at the time of admission did not identify the index patients as suspect EVD cases. Screening children may be challenging because of difficulty eliciting their symptoms and exposure history.
Furthermore, among children in tropical settings there is a high prevalence of other gastrointestinal and febrile illnesses7 that may present similarly to EVD. Among pregnant women, cardinal EVD symptoms, including abdominal pain, vomiting, and muscle and joint pains, might be mistaken for normal pregnancy or complications of labor. Screening admitted patients on the ward every shift or at least 3 times per day as is now recommended by the Sierra Leone Ministry of Health and Sanitation (MoHS)8 is essential for identifying patients who may have subsequently developed signs and symptoms consistent with EVD. Because of the lack of routine EVD screening on the general wards, neither index patient was appropriately isolated at the onset of EVD symptoms, resulting in nosocomial transmission and subsequent infection in 1 health care worker, 1 cleaner, and 4 patients and visitors and interruption of health care services due to staff quarantine and outbreak investigation activities. Aggregate exposure data from both outbreaks demonstrate that high-risk exposures that increase the likelihood for contact with
272
A.C. Dunn et al. / American Journal of Infection Control 44 (2016) 269-72
body fluids (eg, performing exams, taking vital signs, cleaning body fluid spills or other potentially contaminated surfaces, and performing invasive procedures) in the absence of recommended PPE were commonly reported by health care workers in these facilities. Previous studies have also shown an increased risk in Ebola virus transmission from direct skin contact even after controlling for exposure to bodily fluids,9,10 highlighting the importance prompt isolation of suspected EVD patients and proper PPE use at all times. Two infected contacts, a patient who shared the ward and latrine and a family caregiver, did not report exposure to bodily fluids, but did report direct contact with the skin of an EVD patient. However, it is unknown if any of the identified contacts were exposed to other EVD patients apart from the index cases described. Further investigation is needed to determine if pediatric and maternity patients should be screened using a modified case definition for EVD. In the interim, Sierra Leone MoHS recommends that admitted general ward patients should be rescreened for EVD every shift or at least three times per day. In addition, Sierra Leone MoHS recommends that patients meeting the EVD case definition should be immediately isolated, protecting health care providers and patients, while still meeting the suspect EVD patient’s health care needs. It is important that health workers adhere to standard precautions when interacting with all patients and follow World Health Organization-recommended PPE guidelines while caring for patients where risk of exposure to bodily fluids is likely, regardless of their EVD status.4 To provide essential health services and improve the safety of health care staff and patients in non-EVD health care facilities, the Sierra Leone MoHS has approved standard operating procedures for the provision of health services during the EVD outbreak.8,11 Implementation of effective infection prevention and control practices during the EVD outbreak, ensuring access to recommended PPE, and training of health care providers and staff, are essential components to reduce nosocomial transmission of EVD in non-EVD care facilities.
References 1. Kilmarx P, Clarke K, Dietz P, Hamel MJ, Husain F, McFadden JD, et al. Ebola virus disease in health care workers e Sierra Leone, 2014. MMWR Morb Mortal Wkly Rep 2014;63:1168-71. 2. World Health Organization. Laboratory Diagnosis of Ebola Virus Disease. Geneva, Switzerland: World Health Organization. Available from: http://apps. who.int/iris/bitstream/10665/134009/1/WHO_EVD_GUIDANCE_LAB_14.1_eng. pdf; September 2014. Accessed October 15, 2015. 3. Sierra Leone Ministry of Health and Sanitation. Infection Control and Screening and Isolation of Suspected Ebola Patients at the Peripheral Health Units. Infection Control Guidelines and Training Manual. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; 2014. 4. World Health Organization. Interim Infection Prevention and Control Guidance for Care of Patients with Suspected or Confirmed Filovirus Hemorrhagic Fever in Health-Care Settings, with Focus on Ebola. Geneva, Switzerland: World Health Organization. Available from: http://www.who.int/csr/resources/publi cations/ebola/filovirus_infection_control/en; December 2014. 5. Sierra Leone Emergency Management Program Standard Operating Procedure for Safe, Dignified Medical Burials. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; October 2014. 6. Pathmanathan I, O’Connor K, Adams M, Rao CY, Kilmarx PH, Park BJ, et al. Rapid assessment of Ebola infection prevention and control needs - six districts, Sierra Leone, October 2014. MMWR Morb Mortal Wkly Rep 2014; 63:1172-4. 7. Mupere E, Kaducu OF, Yoti Z. Ebola haemorrhagic fever among hospitalised children and adolescents in nothern Uganda: epidemiologic and clinical observations. Afr Health Sci 2001;1:60-5. 8. Government of Sierra Leone. Sierra Leone Emergency management Program standard operating procedures for the Safe provision of hospital services during a haemorrhagic fever outbreak, with a focus on ebola. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; February 2015. 9. Dowell SF, Mukunu R, Ksiazek TG, Khan AS, Rollin PE, Peters CJ. Transmission of ebola Hemorrhagic fever: a Study of risk factors in family members, Kikwit, Democratic Republic of the Congo, 1995. J Infect Dis 1999; 179(Suppl 1):S87-91. 10. Zaki SR, Shieh W, Greer PW, Goldsmith CS, Ferebee T, Katshitshi J, et al. A novel immunohistochemical assay for the detection of Ebola virus in skin: implications for diagnosis, spread, and surveillance of ebola hemorrhagic fever. J Infect Dis 1999;179(Suppl 1):S36-47. 11. Sierra Leone Emergency Management Program Standard Operating Procedures for Screening and Infection Control of Ebola Virus at PHUs and other Non-Ebola Healthcare Facilities. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; November 2014.
Contents lists available at ScienceDirect
American Journal of Infection Control
American Journal of Infection Control
journal homepage: www.ajicjournal.org
Major article
Nosocomial transmission of Ebola virus disease on pediatric and maternity wards: Bombali and Tonkolili, Sierra Leone, 2014 Angela C. Dunn MD a, b, *, Tiffany A. Walker MD a, b, John Redd MD b, David Sugerman MD b, Jevon McFadden MD b, Tushar Singh MD a, b, Joseph Jasperse MPH c, Brima Osaio Kamara MD d, Tom Sesay MD e, James McAuley MD b, Peter H. Kilmarx MD b a
Epidemic Intelligence Service, Centers for Disease Control and Prevention, Atlanta, GA CDC Sierra Leone Ebola Response Team, Freetown, Sierra Leone Concern Worldwide, Tonkolili District, Sierra Leone d Sierra Leone Ministry of Health and Sanitation, Tonkolili District, Sierra Leone e Sierra Leone Ministry of Health and Sanitation, Bombali District, Sierra Leone b c
Key Words: West Africa Hospital-acquired infection Infection control Epidemiology Outbreak
Background: In the largest Ebola virus disease (EVD) outbreak in history, nosocomial transmission of EVD increased spread of the disease. We report on 2 instances in Sierra Leone where patients unknowingly infected with EVD were admitted to a general hospital ward (1 pediatric ward and 1 maternity ward), exposing health care workers, caregivers, and other patients to EVD. Both patients died on the general wards, and were later confirmed as being infected with EVD. We initiated contact tracing and assessed risk factors for secondary infections to guide containment recommendations. Methods: We reviewed medical records to establish the index patients’ symptom onset. Health care workers, patients, and caregivers were interviewed to determine exposures and personal protective equipment (PPE) use. Contacts were monitored daily for EVD symptoms. Those who experienced EVD symptoms were isolated and tested. Results: Eighty-two contacts were identified: 64 health care workers, 7 caregivers, 4 patients, 4 newborns, and 3 children of patients. Seven contacts became symptomatic and tested positive for EVD: 2 health care workers (1 nurse and 1 hospital cleaner), 2 caregivers, 2 newborns, and 1 patient. The infected nurse placed an intravenous catheter in the pediatric index patient with only short gloves PPE and the hospital cleaner cleaned the operating room of the maternity ward index patient wearing short gloves PPE. The maternity ward index patient’s caregiver and newborn were exposed to her body fluids. The infected patient and her newborn shared the ward and latrine with the maternity ward index patient. Hospital staff members did not use adequate PPE. Caregivers were not offered PPE. Conclusions: Delayed recognition of EVD and inadequate PPE likely led to exposures and secondary infections. Earlier recognition of EVD and adequate PPE might have reduced direct contact with body fluids. Limiting nonhealth-care worker contact, improving access to PPE, and enhancing screening methods for pregnant women, children, and inpatients may help decrease EVD transmission in general health care settings. Published by Elsevier Inc. on behalf of the Association for Professionals in Infection Control and Epidemiology, Inc.
* Address correspondence to Angela C. Dunn, MD, Bureau of Epidemiology, Utah Department of Health, PO Box 142104, Salt Lake City, UT 84114-2104. E-mail address: [email protected] (A.C. Dunn). ACD and TAW contributed equally to this work. The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the Centers for Disease Control and Prevention or the institutions with which the authors are affiliated. Conflicts of interest: None.
During the largest Ebola (EVD) outbreak in history, health care workers became infected with EVD while delivering patient care. As of October 31, 2014, 199 of 3,854 (5.2%) laboratory-confirmed EVD cases reported from Sierra Leone were in health care workers.1 We report on 2 patients who were admitted to and died on general hospital wards in October 2014 and were confirmed postmortem as being EVD positive via oral swab by the Centers for Disease Control and Prevention (CDC) Sierra Leone EVD laboratory,2 potentially exposing health care workers to EVD. In Bombali
0196-6553/$00.00 - Published by Elsevier Inc. on behalf of the Association for Professionals in Infection Control and Epidemiology, Inc. http://dx.doi.org/10.1016/j.ajic.2015.09.016
270
A.C. Dunn et al. / American Journal of Infection Control 44 (2016) 269-72
District, a child was admitted to a pediatric ward for presumptive malaria and in Tonkolili District, a pregnant woman was admitted to a maternity ward for prolonged labor. Both patients were screened for EVD on admission; however, neither patient was reassessed for symptoms of EVD after admission. Both patients later died from undiagnosed EVD on the general wards, confirmed by postmortem testing. Both hospitals used the Sierra Leone Ministry of Health and Sanitation EVD case definition to triage patients on admission. The Sierra Leone EVD case definition defines a suspected EVD patient as any person with fever (>38 C) plus 3 or more of the following symptoms: vomiting, headache, nausea, diarrhea, difficulty breathing, fatigue, abdominal pain, loss of appetite, muscle or joint pain, unexplained bleeding, difficulty swallowing, or hiccups. Or, a suspected EVD patient is any ill person with an epidemiologic link to a suspected or confirmed EVD case.3 The CDC assisted the District Health Management Teams in investigating these health care-related EVD virus exposures. Those who had direct contact with EVD-infected patients were monitored for signs and symptoms of EVD over the 21-day incubation period. EVD was diagnosed at the CDC-Sierra Leone EVD laboratory by testing serum using reverse transcriptionepolymerase chain reaction (RT-PCR).2 Occupational exposures of those who became infected with EVD in this cohort included cleaning body fluids and placing intravenous catheters. It is essential that health care workers adhere to standard precautions3,4 when interacting with all patients. Both health care workers and caregivers need access to recommended personal protective equipment (PPE) and training on proper use. During an EVD outbreak, the following might help decrease EVD transmission in general health care settings: adhering to World Health Organization-recommended PPE guidelines, implementing effective EVD screening methods, and reassessing admitted patients for EVD. BOMBALI DISTRICT On October 8, 2014, a 29-month-old child presented with vomiting and fatigue to the pediatric outpatient center of the district’s government hospital. The patient did not meet the EVD screening case definition and therefore was admitted to the general pediatric ward, where he was treated for malaria. The patient developed diarrhea later that day. On October 10, he developed fever, rash, red eyes, and melena. The patient died October 13 on the pediatric ward. Given the patient’s symptoms, the attending pediatrician requested a serum specimen be drawn for EVD RT-PCR at the time of death and the results confirmed EVD. A total of 39 health care workers had contact with the source patient, including 32 nurses, 3 cleaners, 2 laboratory technicians, 1 mortuary worker, and 1 porter. CDC performed a review of staffing records and conducted in-person interviews with 28 (72%) of the health care worker contacts to determine the dates and circumstances of their exposure to the index patient. Thirty-six (92%) health care worker contacts lacked access to proper PPE; however, all 3 cleaners had access to and wore World Health Organizationrecommended PPE.3 There were 11 (31%) exposed nurses who were unavailable for interview due to author’s time constraints. A summary of exposures is presented in Table 1. At the direction of the hospital’s medical superintendent, the 36 health care workers with confirmed direct contact were placed on home quarantine for 21 days. The 3 cleaners who wore adequate PPE were excluded from home quarantine. A line list of exposed health care workers was submitted to the District Health Management Office. Contact tracers were assigned to each health care worker for daily direct monitoring of signs and symptoms of EVD. Those who developed symptoms were isolated in a designated room at the district’s government hospital. Their blood
specimens were sent to the CDC-Sierra Leone EVD Laboratory for testing. Of the 36 contacts, 1 health care worker developed symptoms 7 days after the occupational exposure and tested positive for EVD 6 days later, resulting in a health care worker attack rate of 2.8%. The infected individual was a nurse whose only reported exposure was placing an intravenous catheter in the index patient while wearing short gloves only. Exposure and outcome data for this nurse are presented in Table 2. TONKOLILI DISTRICT A 25-year-old full term pregnant woman presented with labor pains to her community maternity clinic on October 7, 2014. On October 9, she was transferred to the maternity ward of a general hospital due to prolonged labor where she was observed for 5 days. On October 14, the patient gave birth to a live newborn via an uncomplicated cesarean section for prolonged labor. Two days later, on October 16, she reported feeling febrile and had 1 episode of vomiting. The following day, October 17, the patient developed red eyes and had excessive bleeding from her surgical site and vagina. The physician noted in the patient’s chart that he would discuss isolation with the care team; meanwhile, the patient remained on the general maternity ward. The patient passed away on the general maternity ward October 18. Consistent with Sierra Leone national protocol,5 a postmortem buccal swab for EVD RT-PCR was performed and the results confirmed EVD. At the direction of the hospital’s medical superintendent and the district medical officer, all hospital staff, patients, and caregivers who came into direct contact with the index patient were quarantined in the maternity ward for 21 days. The medical superintendent assigned 1 staff member to screen the quarantined staff twice daily, including monitoring symptoms and checking body temperatures using an infrared thermometer. Screening data were recorded in a logbook that was reviewed in the mornings and evenings by the medical superintendent. The medical superintendent took the temperatures and assessed symptoms of the quarantined patients and their caregivers twice daily. Those persons who developed symptoms consistent with EVD were isolated in a predetermined, individual hospital room. Their blood specimens were sent to the CDC-Sierra Leone EVD Laboratory for testing. In addition, the district medical officer placed the contacts from the community maternity clinicd3 nursesdon home quarantine and active surveillance with daily assessments by a contact tracer for 21 days. A total of 46 people reportedly came into contact with the index patient from the time of her presentation to the community maternity clinic on October 7, including 25 hospital staff: 18 nurses, 5 hospital cleaners, 1 physician, and 1 laboratory technician. In addition, 7 family caregivers, 4 adult patients, 4 newborns, 3 children, and 3 community maternity clinic nurses were exposed. We reviewed medical records of the index patient to establish the timeline of events, and conducted in-person interviews with 37 contacts (80%) to determine their exposure history. Maternity ward nurses and hospital cleaning staff lacked access to recommended PPE and caregivers were not offered PPE. Out of the 8 exposed persons (10%) who were unable to be interviewed, 6 were children, 1 was the caregiver of the index patient who was at an EVD treatment unit with the index patient’s newborn, and 2 were hospital staff who were unable to be located. Maternity ward health care workers provided exposure history for these 8 contacts. A summary of exposures is presented in Table 1. Of the 46 contacts, 6 tested positive for EVD within 21 days of exposure to the index patient, resulting in a health care facility attack rate of 13.0%. Those who became infected were the newborn and caregiver of the index patient, 1 maternity ward patient and her newborn and caregiver,
A.C. Dunn et al. / American Journal of Infection Control 44 (2016) 269-72
271
Table 1 Exposures to patients with Ebola virus disease (EVD) and personal protective equipment (PPE) worn by contacts while exposed to patients with EVD, Bombali and Tonkolili Districts, Sierra Leone, October 2014
Exposure
Contacts with exposure (N ¼ 82)*
PPE worn while exposed
Shared ward/latrine Newborn of index EVD patient Newborn of confirmed EVD maternity ward patient Contact with newborn of an EVD patient Took vital signs Cleaned linens Vaginal exam Performed/assisted in cesarean section Cleaned body fluids Cleaned body fluids Cleaned surfaces: floor, walls, bed Cleaned surgical instruments Moved patient Placed urinary catheter Placed intravenous line Gave intravenous medications Blood draw Discontinued intravenous line Gave intramuscular medications Changed surgical site dressing General touching patient General touching patient Death pronouncement Unknown contact No known direct contact
None None None None Short gloves Short gloves Short gloves Gown; short gloves (three pairs); mask; goggles; shoe covers Short gloves None Gown; apron; short gloves (2 pairs); mask Short gloves Short gloves Short gloves; gown Short gloves Short gloves; gown Gown; apron; short gloves (2 pairs); mask Short gloves Short gloves Short gloves; gown Short gloves None Short gloves Unknown None
15 1 1 4 6 2 8 3 4 1 3 1 4 1 9 15 4 1 1 3 5 4 1 13 4
(18) (1) (1) (5) (7) (2) (10) (4) (5) (1) (4) (1) (5) (1) (11) (18) (5) (1) (1) (4) (6) (4) (1) (16) (5)
Contacts with exposure who developed EVD* 3 1 1 2 1
1 1
1 1
2
(20) (100) (100) (50) 0 (50) 0 0 (25) (100) 0 0 (25) 0 (11) 0 0 0 0 0 0 (50) 0 0 0
NOTE. Values are presented as n (%). *Some contacts had multiple exposures.
Table 2 Exposures and outcomes of contacts who developed Ebola virus disease, Bombali and Tonkolili Districts, Sierra Leone, October 2014 Contact ID
District
Relationship with index patient
1 2
Bombali Tonkolili
Nurse Newborn
3 4
Tonkolili Tonkolili
5*
Tonkolili
Caregiver Shared ward and latrine Shared ward
6*
Tonkolili
7
Tonkolili
Shared ward and latrine Hospital cleaner
Exposure(s)
Date(s) of exposure(s) to index patient
PPE worn while exposed
Date of symptom onset
Outcome
Placed intravenous catheter Intrauterine, cesarean section, breast milk, body contact Cleaned body fluids Body contact
October 10 Antenatal, October 14-18 October 9-18 October 9-18
Short gloves None
October 17 October 17
Death on October 27 Death on October 24
None None
November 1 October 22
Death on November 10 Death on October 28
Newborn of contact 4
Antenatal; October 16-18 October 9-18
None
October 23
Death on unknown date
None
Unknown
Death on unknown date
October 14
Short gloves
October 22
Death on unknown date
Body contact, caregiver for contact 4 Cleaned body fluids; moved patient; cleaned linens
*Contacts 5 and 6 might have contracted Ebola virus disease from contact 4 rather than the index patient, and thus might be secondary cases.
and 1 hospital cleaner. The newborn of the index patient was likely infected via mother-to-child transmission. The newborn of the infected maternity ward patient might have been infected via his/her mother’s breast milk. A summary of exposures and outcomes of those who contracted EVD is presented in Table 2. DISCUSSION Unidentified EVD patients in non-EVD health care facilities present a significant health risk to health care workers, caregivers, and other patients.6 Effective infection prevention and control measures to reduce the risk of nosocomial transmission to health care workers and patients include improved disease screening, isolation of suspect case patients, facility engineering controls, and adherence to recommended PPE.4 In the outbreaks we describe, screening at the time of admission did not identify the index patients as suspect EVD cases. Screening children may be challenging because of difficulty eliciting their symptoms and exposure history.
Furthermore, among children in tropical settings there is a high prevalence of other gastrointestinal and febrile illnesses7 that may present similarly to EVD. Among pregnant women, cardinal EVD symptoms, including abdominal pain, vomiting, and muscle and joint pains, might be mistaken for normal pregnancy or complications of labor. Screening admitted patients on the ward every shift or at least 3 times per day as is now recommended by the Sierra Leone Ministry of Health and Sanitation (MoHS)8 is essential for identifying patients who may have subsequently developed signs and symptoms consistent with EVD. Because of the lack of routine EVD screening on the general wards, neither index patient was appropriately isolated at the onset of EVD symptoms, resulting in nosocomial transmission and subsequent infection in 1 health care worker, 1 cleaner, and 4 patients and visitors and interruption of health care services due to staff quarantine and outbreak investigation activities. Aggregate exposure data from both outbreaks demonstrate that high-risk exposures that increase the likelihood for contact with
272
A.C. Dunn et al. / American Journal of Infection Control 44 (2016) 269-72
body fluids (eg, performing exams, taking vital signs, cleaning body fluid spills or other potentially contaminated surfaces, and performing invasive procedures) in the absence of recommended PPE were commonly reported by health care workers in these facilities. Previous studies have also shown an increased risk in Ebola virus transmission from direct skin contact even after controlling for exposure to bodily fluids,9,10 highlighting the importance prompt isolation of suspected EVD patients and proper PPE use at all times. Two infected contacts, a patient who shared the ward and latrine and a family caregiver, did not report exposure to bodily fluids, but did report direct contact with the skin of an EVD patient. However, it is unknown if any of the identified contacts were exposed to other EVD patients apart from the index cases described. Further investigation is needed to determine if pediatric and maternity patients should be screened using a modified case definition for EVD. In the interim, Sierra Leone MoHS recommends that admitted general ward patients should be rescreened for EVD every shift or at least three times per day. In addition, Sierra Leone MoHS recommends that patients meeting the EVD case definition should be immediately isolated, protecting health care providers and patients, while still meeting the suspect EVD patient’s health care needs. It is important that health workers adhere to standard precautions when interacting with all patients and follow World Health Organization-recommended PPE guidelines while caring for patients where risk of exposure to bodily fluids is likely, regardless of their EVD status.4 To provide essential health services and improve the safety of health care staff and patients in non-EVD health care facilities, the Sierra Leone MoHS has approved standard operating procedures for the provision of health services during the EVD outbreak.8,11 Implementation of effective infection prevention and control practices during the EVD outbreak, ensuring access to recommended PPE, and training of health care providers and staff, are essential components to reduce nosocomial transmission of EVD in non-EVD care facilities.
References 1. Kilmarx P, Clarke K, Dietz P, Hamel MJ, Husain F, McFadden JD, et al. Ebola virus disease in health care workers e Sierra Leone, 2014. MMWR Morb Mortal Wkly Rep 2014;63:1168-71. 2. World Health Organization. Laboratory Diagnosis of Ebola Virus Disease. Geneva, Switzerland: World Health Organization. Available from: http://apps. who.int/iris/bitstream/10665/134009/1/WHO_EVD_GUIDANCE_LAB_14.1_eng. pdf; September 2014. Accessed October 15, 2015. 3. Sierra Leone Ministry of Health and Sanitation. Infection Control and Screening and Isolation of Suspected Ebola Patients at the Peripheral Health Units. Infection Control Guidelines and Training Manual. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; 2014. 4. World Health Organization. Interim Infection Prevention and Control Guidance for Care of Patients with Suspected or Confirmed Filovirus Hemorrhagic Fever in Health-Care Settings, with Focus on Ebola. Geneva, Switzerland: World Health Organization. Available from: http://www.who.int/csr/resources/publi cations/ebola/filovirus_infection_control/en; December 2014. 5. Sierra Leone Emergency Management Program Standard Operating Procedure for Safe, Dignified Medical Burials. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; October 2014. 6. Pathmanathan I, O’Connor K, Adams M, Rao CY, Kilmarx PH, Park BJ, et al. Rapid assessment of Ebola infection prevention and control needs - six districts, Sierra Leone, October 2014. MMWR Morb Mortal Wkly Rep 2014; 63:1172-4. 7. Mupere E, Kaducu OF, Yoti Z. Ebola haemorrhagic fever among hospitalised children and adolescents in nothern Uganda: epidemiologic and clinical observations. Afr Health Sci 2001;1:60-5. 8. Government of Sierra Leone. Sierra Leone Emergency management Program standard operating procedures for the Safe provision of hospital services during a haemorrhagic fever outbreak, with a focus on ebola. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; February 2015. 9. Dowell SF, Mukunu R, Ksiazek TG, Khan AS, Rollin PE, Peters CJ. Transmission of ebola Hemorrhagic fever: a Study of risk factors in family members, Kikwit, Democratic Republic of the Congo, 1995. J Infect Dis 1999; 179(Suppl 1):S87-91. 10. Zaki SR, Shieh W, Greer PW, Goldsmith CS, Ferebee T, Katshitshi J, et al. A novel immunohistochemical assay for the detection of Ebola virus in skin: implications for diagnosis, spread, and surveillance of ebola hemorrhagic fever. J Infect Dis 1999;179(Suppl 1):S36-47. 11. Sierra Leone Emergency Management Program Standard Operating Procedures for Screening and Infection Control of Ebola Virus at PHUs and other Non-Ebola Healthcare Facilities. Freetown, Sierra Leone: Sierra Leone Ministry of Health and Sanitation; November 2014.
