Secondary Logo

Journal Logo

Global Health: Research Report

Anesthesia Practice and Perioperative Outcomes at Two Tertiary Care Hospitals in Freetown, Sierra Leone

Koka, Rahul MD, MPH*; Chima, Adaora M. MBBS, MPH*; Sampson, John B. MD*; Jackson, Eric V. MD, MBA*; Ogbuagu, Onyebuchi O. MBBS, MPH*; Rosen, Michael A. PhD*†; Koroma, Michael MBBS; Tran, Tina P. MD*; Marx, Megan K. BSN, RN, MPH*; Lee, Benjamin H. MD, MPH*

Author Information
doi: 10.1213/ANE.0000000000001285
  • Free

In 2008, the global volume of surgery was estimated to be 234 million major operations per year, or roughly 1 operation for every 25 persons.1 Only 8.1 million, or 3.5% of these procedures, occur in the poorest third of the world’s population—in countries with an annual per-head expenditure on health ≤$100. Perioperative mortality is estimated to be 5% to 10% in these regions.1 More recently, research has focused on the role of operative personnel, infrastructure, and training in tackling the global surgical disease burden. However, current studies on perioperative outcomes in resource-constrained countries are often retrospective reviews of operating room (OR) log books and records that likely do not capture the details of anesthesia care practice and intraoperative events. In this study, we examined anesthesia practice in Sierra Leone through direct observation of perioperative tasks, events, and outcomes.

The Republic of Sierra Leone is a country of 5.7 million people located on the West African coast.2 An 11-year-long civil war (1991–2002) resulted in the death of an estimated 70,000 people, the displacement of one-third of the country’s population, and the destruction of infrastructure, including hospitals and the health workforce.3,4 Unfortunately, although recovery of the country’s health system has been steady, it has been inhibited by a paucity of resources, a high poverty index, and the overwhelming magnitude of the human and infrastructure loss recently compounded by an ongoing Ebola hemorrhagic fever outbreak. In 2010, the life expectancy at birth was 57.4 years (rated 201/223 countries worldwide), the maternal mortality rate was 890 per 100,000 births (fifth highest in the world), and the under-5 (U5) mortality rate was 276 deaths per 1000 (highest in the world). In developing countries, many causes of maternal and childhood morbidity and mortality are surgical. In a 2012 survey of the prevalence of surgical conditions and surgically treatable deaths in Sierra Leone, 25% of conditions that respondents reported as needing surgical attention, and 25% of deaths in household members might have been averted by timely surgical care.3

In addition to having a very low physician density (0.2/1000—ranked 190/192 in 2008),2 the government hospitals in Sierra Leone, which provide most of the surgical care for the country’s population, have severe shortages in infrastructure and supplies.4,5 Sierra Leone has 3 government-run tertiary hospitals that also serve as teaching hospitals: Connaught Hospital, Princess Christian Maternity Hospital (PCMH), and Ola During Children’s Hospital, all located in Freetown. Surgical and anesthesia care is delivered in the former 2 hospitals.

The goal of this study was to comprehensively examine anesthesia practice in a resource-constrained setting to identify threats and hazards in care that might be addressed with contextually appropriate interventions. The anesthesia providers in Connaught Hospital and PCMH represent roughly 25% of Sierra Leone’s public anesthesia workforce. Therefore, an examination of practice in these institutions will likely reflect the surgical and anesthesia care accessible to most of the country’s population.


Study Design

After receiving approval from the Sierra Leone Scientific Review and Ethics Committee and the Johns Hopkins institutional review board, we conducted a prospective observational study of anesthesia practice at 2 study sites. Anesthesia data records were designed and piloted by anesthesia providers at the Johns Hopkins Hospital and modified through a piloting process in Sierra Leone to ensure contextually appropriate data variables and minimize ambiguity in the data collection processes. Data collectors underwent 6 weeks of supervised training in the OR, where they observed and documented anesthesia tasks and operating conditions that could influence a change in anesthetic delivery. They were supervised during the process until a minimum kappa statistic of 0.7 was attained.

Study Setting

This study took place at PCMH and Connaught Hospital, 2 tertiary, government-supported hospitals in Freetown, Sierra Leone. PCMH is the national maternal referral hospital, providing only obstetric and gynecologic services. Connaught Hospital is the national referral hospital for all health services except maternal and nonsurgical pediatric health care. We collected data at both hospitals over a period of 8 months, from June 2012 to February 2013.


All anesthesia providers at both hospitals were eligible for this study. Participants volunteered, and informed consent was obtained.

Variables and Data Measurement

General information on the hospitals and their infrastructure was collected using the World Health Organization Global Initiative for Emergency and Essential Surgical Care assessment toolkit and a modified version of a hospital questionnaire developed by Jochberger et al.6 These tools were administered to hospital matrons, statisticians, and hospital management officials.

In addition, anesthesia-related tasks were documented from the time of patient contact in the preoperative area to postoperative handoff. We collected information regarding patient age, gender, and pre-existing conditions to contextualize intraoperative vital signs. Other information was obtained from the anesthetic records and operating list or by verbal confirmation from the anesthesia provider.

During the procedure, we documented anesthesia-related tasks such as medications administered and vital signs monitored or recorded by the anesthetists in the OR. We also noted performance time of events such as anesthesia start and stop times as well as electrical and mechanical disturbances during the operation (Appendix 1). After surgery, we made inquiries about the status of admitted patients daily for up to 30 days to determine discharge date, returns to the OR, or patient death.

Statistical Methods

Anesthesia records and follow-up data were scanned, and abstracted data were entered into a File MakerPro 12 database (FileMaker, Inc., Santa Clara, CA). STATA 11 software (StataCorp LP, College Station, TX) and Microsoft Excel (Microsoft Corp., Redmond, WA) were used to analyze the data. Perioperative details regarding observed cases were reported as frequency distributions.

Role of the Funding Source

The study sponsor, Gradian Health Systems LLC, played no role in the study design, data collection, analysis or interpretation, writing of the manuscript, or decision to submit this report.


Hospital Resources and Infrastructure

Table 1.
Table 1.:
World Federation of Societies of Anesthesiologists International Standards Assessment at the 2 Study Sites

Hospital infrastructure and resources are outlined in Table 1. We observed that both hospitals suffered from electrical supply interruptions, and only Connaught Hospital had a consistent water supply. Neither hospital had wall oxygen. Anesthesia medications were obtained through the medical stores of the Ministry of Health and Sanitation, donations from medical missions, and (at Connaught Hospital) a revolving drug fund managed by the hospital anesthesia team. PCMH had no functioning anesthesia machine at the time of this study. Neither hospital had a system for scavenging waste anesthetic gases, and each hospital had 1 physician anesthetist in attendance.


All 25 nurse anesthetists and 2 physician anesthesiologists volunteered for this study. Data collectors observed 754 anesthetics: 504 at Connaught Hospital and 250 at PCMH. The observed cases represent approximately 59% of all cases performed at Connaught and 29% of cases performed at PCMH within the 8-month study period.

Preoperative Assessment

Table 2.
Table 2.:
Description of Patient Demographics and Observed Preoperative Assessments

Observed preoperative tasks such as obtaining consent and medical histories are listed in Table 2. Medical history was documented on patient charts approximately half the time, and the most commonly documented diseases were hypertension (6.8%), diabetes mellitus (3.2%), and asthma (1.2%).

Intraoperative Findings

Tables 3 and 4 list the 10 most common operative procedures observed at each facility. Anesthetic technique varied even with the same operative procedure. Only 4 anesthetic techniques were observed: general anesthesia, regional anesthesia (spinal only), monitored anesthesia care, and local medication given by the surgeon (Table 5). The number of general anesthetics performed includes the unplanned conversions to general anesthesia from regional anesthesia when injection into the subarachnoid space was unsuccessful or analgesic effects terminated before the conclusion of surgery (25 cases at Connaught and 37 cases at PCMH). At Connaught Hospital, 25 of 163 (15.3%) attempted spinal anesthetics were converted to general anesthetic; the incidence of similar conversions at PCMH was 37 of 200 (18.5%). However, at Connaught, when spinal anesthesia failed, only 12% were converted to general anesthesia with an endotracheal tube compared with 2.7% (1 of 37) at PCMH, where all patients then received total IV anesthetic and required manual bag-valve-mask ventilation.

Table 3.
Table 3.:
Anesthesia Techniques Performed for 10 Most Common Surgical Operations at Connaught Hospital
Table 4.
Table 4.:
Anesthesia Techniques Performed for 10 Most Common Surgical Operations at Princess Christian Maternity Hospital
Table 5.
Table 5.:
Intraoperative Anesthesia Procedures and Medications

Halothane was the only inhalational anesthetic available, bupivacaine the sole spinal anesthetic, and lignocaine the sole local anesthetic available. PCMH was unable to deliver inhalational anesthesia; therefore, all general anesthetics were performed using IV ketamine (doses ranged from 50 to 900 mg), thiopentone, or propofol (Table 5).

Table 6.
Table 6.:
Intraoperative Events (Clinical and Nonclinical)

Intraoperative vital sign monitoring depended on the patient’s age and the availability of functional monitors and power supply. Power outage occurred during 7.8% of observed cases (Table 6). Both hospitals were limited in the number of intraoperative monitors available, and their utility varied among patient types. At Connaught Hospital, blood pressure was measured in 35.9% (52/145) of pediatric cases but in 90.5% (325/359) of adult cases. In contrast, pulse oximetry was monitored in 95.2% of children and 86.1% of adults. At PCMH, blood pressure and pulse oximetry were monitored in 98% and 88.8% of cases, respectively, but did not seem to relate to age group. Both hospitals reported having a recovery room, although no support staff was available for patient care in these areas. Clinical intraoperative events such as hypotension were noted with frequencies as described in Table 6. There were 37 cases in which significant hypotension was observed, and in these cases, 100% were treated with intravascular fluids, 45.9% with autologous blood transfusions, and 81.1% received vasoactive agents, which included epinephrine, ephedrine, and atropine.

Postoperative Care and Outcomes

The postoperative care period was defined as the time from OR departure to discharge, death, or 30 days, whichever came first. In the immediate postoperative period, 22.7% (171/754) of observed patients were placed in a designated recovery room or area; 52% (90/171) of these cases were pediatric patients with age <18 years and received intermittent clinical assessment by anesthesia providers; 65.1% (491/754) of patients were placed in a hallway outside the OR until a nursing team member was available to transfer the patient to the ward; 0.4% (3/754) had immediate ward transfers; 0.4% (3/754) were transferred directly to the high-dependency unit; and 11.4% (86/754) were either discharged home immediately or could not be traced in the immediate postoperative period.

At Connaught Hospital, the 30-day in-hospital mortality rate was 1.4% (7/504) among observed cases and the 30-day in-hospital mortality rate at PCMH was 0.8% (2/250) with mortality in pregnancy-related conditions occurring in 0.5% (1/210). Thirty cesarean deliveries resulted in stillbirth deliveries, 6.7% (2/30) of which were documented preoperatively.

Postoperative vital sign monitoring was performed in 17.1% (129/754) of all cases; 96.9% (125/129) of which were pulse oximetry assessments that identified 9 cases of hypoxemia. The postoperative status of 51 cases (6.7%) could not be verified owing to inconclusive ward documentation. The composition of anesthesia technique and case type in these cases did not vary significantly from the remainder of the observed cohort.


Our study captured anesthesia delivery practice by 25% of the Sierra Leonean anesthesia workforce stationed in these hospitals, which serve as teaching hospitals for anesthesia trainees and the sole medical school in the country. Nurse anesthetists currently make up approximately 98% of anesthesia providers in the Sierra Leonean public health sector and performed 91% of the cases observed with minimal or no direct physician oversight. Nurse anesthetists in Sierra Leone go through a United Nations Population Fund-sponsored 12-month anesthesia training program with prerequisite training in nursing (3-year program) and midwifery (2.5 years). They also receive annual update courses coordinated by the Sierra Leone Association of Nurse Anesthetists with the support of the United Nations Population Fund and other nongovernmental organizations.

World Federation of Societies of Anesthesiologists Standards and Procedures Performed

In 2010, the World Federation of Societies of Anesthesiologists (WFSA) published standards intended to improve the safe practice of anesthesia worldwide, even in resource-limited settings.7 These guidelines outline “suggested,” “recommended,” and “highly recommended” aspects of hospital infrastructure, supplies, and professional standards and provide a designation of 3 levels of hospital care: small hospital, district/provincial hospital, and referral (tertiary-care) hospital. Each level is expected to exceed the standards outlined for the preceding level. We assessed these essential elements and compared those of the study hospitals with the WFSA standards/guidelines (Table 1). Both hospitals provide some degree of level 3 care, because they are considered tertiary-care, referral hospitals in the governmental hospital system. They also provide many surgical services described for tertiary-level hospitals. However, by the size and resources available, the sites more closely resemble WFSA level 2 facilities.

Available Resources and Perioperative Practice

Anesthesia practice is influenced by available resources, equipment, and comfort of the provider. Thus, standards and protocols will differ between hospitals, even within the same cities. It is not surprising, therefore, that regional anesthesia was preferred at PCMH (65.2%) and general anesthesia was the preferred technique at Connaught Hospital (56.9%; Table 5). Although several anesthesia delivery systems were on-site at both hospitals, none was functional at PCMH. At Connaught Hospital, a Glostavent® (Diamedica Ltd., UK) with a draw-over system, which had a nonfunctional ventilator and oxygen concentrator, was retired in the early stages of the study because of disrepair and a Compact-3 (a type of Boyle machine) that was often in different stages of repair was in intermittent use during the course of the study. Compressed oxygen was unavailable at PCMH for the course of the study, and supplemental oxygen was delivered through an oxygen concentrator. At Connaught Hospital, a concentrator was used in 55.2% (186/337) of all cases that required supplemental oxygen; 95 of such patients were also ventilated with an anesthesia machine. The anesthesia machine, which is designed to be used with an oxygen cylinder or pipeline oxygen, had been adapted for use with a standalone oxygen concentrator although neither the concentrator nor the machine was used with an oxygen analyzer to measure fraction of inspired oxygen. Moreover, because oxygen concentrators require electricity, the frequent power outages observed present a potential hazard to patients who require consistent intraoperative oxygen supplementation.

WFSA-recommended practice on preoperative anesthesia evaluation is a comprehensive assessment and documentation of relevant findings on the patient’s status. In this study, we documented preoperative reviews that were recorded on file or witnessed by the data collector. Although previous day clinical review was not routine practice at these hospitals, such interactions, if they occurred, were not captured in this study if undocumented in the patient’s chart. Infrequent monitoring in the postoperative period could be partially attributed to the allocation of available monitoring devices to the OR.

Narcotic medications are highly regulated in Sierra Leone, and this likely impacts its clinical use, as seen in this study, in which 3.6% (27/754) of cases were administered any form of narcotic (morphine or tramadol).


When a general anesthetic was used at PCMH, it was always with anesthetic doses of ketamine (86/86 cases), and only 7 patients of this cohort were tracheally intubated. Like in many parts of the developing world, ketamine was the most commonly used IV anesthetic administered during the study period.8 Ketamine does not depress ventilation, is likely to maintain hemodynamic stability, and has a high margin of safety.9 It is especially favorable for use by those not trained in anesthesia or when basic monitoring equipment is lacking. In the obstetric patient, it can be particularly useful in cases of obstetric hemorrhage; however, anesthetic doses should be used with caution in patients who do not have a definitive airway and in those with pre-eclampsia or hypertension.

Spinal Converted to General Anesthesia

During conversion of spinal to general anesthesia, providers at PCMH were often hesitant to secure the airway by using endotracheal intubation (in patients who are at risk for aspiration pneumonia) for concern of potential airway loss and risks of hypoxia/death. A contributing factor to the frequency of general anesthesia without an endotracheal tube at PCMH may relate to the lack of a functional anesthetic machine there. Although the hospital owned 2 recent-model anesthesia machines, neither was in clinically operational condition because of issues related to lack of compressed gas, absence of carbon dioxide-absorbing granules, and lack of maintenance support.

The providers attributed the failed spinal anesthetics to the following: inability to access the subarachnoid space (in some instances after multiple attempts), experience of the anesthesia provider or the potency of the anesthetic drug, and/or the extended duration (in a few instances) of the surgical cases. A prospective study on failed obstetric spinal anesthesia in Nigeria reported that the technique and level of experience of the anesthesia provider were significantly related to the rate of failure.10 The significant incidence of failed spinal anesthesia indicates a need for reinforced training on spinal anesthesia procedures and endotracheal anesthesia techniques in the urgent/emergent clinical situation. It is interesting that 53 of 54 (98%) cesarean deliveries performed with the patient under IV general anesthesia were conducted without physical airway protection with a cuffed endotracheal tube. A study on obstetric mortality in Nigeria showed that failed airway management remains a significant contributor to anesthesia-related maternal mortality.11

Perioperative Mortality

We estimated that mortality was 8 deaths/1000 anesthetics at PCMH and 13.8 deaths/1000 anesthetics at Connaught Hospital. These rates are comparable with reported mortality rates of countries such as Malawi (10.5 deaths/1000 anesthetics),12 Zimbabwe (2.9 deaths/1000 anesthetics),13 and Togo (20.5 deaths/1000 anesthetics).14 In the United States, anesthesia-related mortality is estimated to be approximately 1.1 death/1,000,000 anesthetics.15

The American Society of Anesthesiologists (ASA) physical status classification system is a proven predictor of postoperative outcomes.16,17 In our study, we observed only 5% of patients who were ASA III to IV compared with 56.5% in the U.S.-based study of outpatient “remote” anesthesia venues.18 It is possible that patients with significant comorbidities have deterrents to presenting to Sierra Leone referral hospitals or are not considered candidates for surgical treatment in low-resource environments. Also, the lower average age of the population and lack of primary medical care may result in lower levels of diagnosed chronic disease. The low numbers of ASA III to IV patients could be responsible for the low rates of morbidity and mortality that we observed.


Our assessment of perioperative mortality was limited by our inability to capture cases that presented outside the normal weekday OR schedule hours (8:30 am to 4:30 pm). Thus, anesthesia procedures performed outside of these hours/days were not observed by the study team and were not included in this analysis. These cases may likely have included emergent cases, which may have been performed under slightly different conditions with potentially different outcomes from the observed cases.


Understanding the current state of anesthesia care practice is an important step in determining appropriate educational and clinical interventions for building anesthesia capacity in any location. It is especially important in challenging clinical environments such as resource-constrained countries that have inconsistent access to supplies and functional equipment, inadequate numbers of trained personnel, and limited opportunities for collaboration and expert consultation. In these locations, specialty-trained nurses often provide anesthesia care to a large segment of the population and will likely continue to do so in the absence of a growing physician workforce.

Our observations identified gaps in the application of internationally recommended safe anesthesia practices. These gaps could be related to a lack of available training and resources such as protocols regarding optimal airway management in obstetric anesthesia delivery.

Impediments to safe anesthesia care delivery can be identified by engaging local providers and stakeholders in the examination of their clinical processes using systematic process improvement techniques such as the Failure Modes and Effects Analysis.19,20 Potential solutions should include designing medical curricula specifically tailored to the needs of the local clinical providers. In addition, in situ, low-cost medical simulation may enhance anesthesia training by targeting performance demonstration and improvement with no risk to patients. Also, anesthesia equipment design and development should consider the characteristics of the environments of end users, the availability of biomedical support, the ease of repair, and the local availability of replacement parts. Anesthesia delivery systems such as the Universal Anaesthesia Machine® (Gradian Health Systems, New York, NY) and the Glostavent® devices, which have draw-over vaporizer systems and inbuilt oxygen concentrators, are specifically designed to overcome some of these issues. Such innovations should be accompanied by local biomedical engineering capacity-building (education, personnel, and infrastructure) to tackle equipment-related issues and improve durability and sustainability.

By thoroughly examining the process of anesthesia delivery in resource-constrained environments, we can better analyze the threats and hazards to care and focus on contextually appropriate interventions to address practice-based systems improvement and relevant curricula for local providers.

This study addresses this need by documenting anesthesia care practice in 2 tertiary hospitals in Freetown, Sierra Leone.



Name: Rahul Koka, MD, MPH.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Rahul Koka has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Rahul Koka received research funding from Gradian Health Systems.

Name: Adaora M. Chima, MBBS, MPH.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Adaora M. Chima has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Adaora M. Chima received research funding from Gradian Health Systems.

Name: John B. Sampson, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: John B. Sampson has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: John B. Sampson received research funding from Gradian Health Systems.

Name: Eric V. Jackson, MD, MBA.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Eric V. Jackson has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Eric V. Jackson received research funding from Gradian Health Systems.

Name: Onyebuchi O. Ogbuagu, MBBS, MPH.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Onyebuchi O. Ogbuagu has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Onyebuchi O. Ogbuagu received research funding from Gradian Health Systems.

Name: Michael A. Rosen, PhD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Michael A. Rosen has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Michael A. Rosen reports grant support from the Agency for Healthcare Research and Quality, The VHA, Inc. Health Research Center, the Patient Centered Outcomes Research Institute, the Gordon and Betty Moore Foundation, and Gradian Health Systems, all for the conduct of patient safety and human factors-related research.

Name: Michael Koroma, MBBS.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Michael Koroma has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Michael Koroma declares no conflicts of interest.

Name: Tina P. Tran, MD.

Contribution: This author helped analyze the data and write the manuscript.

Attestation: Tina P. Tran has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Tina P. Tran declares no conflicts of interest.

Name: Megan K. Marx, BSN, RN, MPH.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Megan K. Marx has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Megan K. Marx received research funding from Gradian Health Systems.

Name: Benjamin H. Lee, MD, MPH.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Benjamin H. Lee has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Conflicts of Interest: Benjamin H. Lee received research funding from Gradian Health Systems.

This manuscript was handled by: Steven L. Shafer, MD.


1. Weiser TG, Regenbogen SE, Thompson KD, Haynes AB, Lipsitz SR, Berry WR, Gawande AAAn estimation of the global volume of surgery: a modelling strategy based on available data.Lancet200837213944
2. The World FactbookAvailable at: Accessed January 13, 2015
3. Groen RS, Samai M, Stewart KA, Cassidy LD, Kamara TB, Yambasu SE, Kingham TP, Kushner ALUntreated surgical conditions in Sierra Leone: a cluster randomised, cross-sectional, countrywide survey.Lancet201238010827
4. Kingham TP, Kamara TB, Cherian MN, Gosselin RA, Simkins M, Meissner C, Foray-Rahall L, Daoh KS, Kabia SA, Kushner ALQuantifying surgical capacity in Sierra Leone: a guide for improving surgical care.Arch Surg20091441227
5. National Health Sector Strategic Plan 2010–2015Available at: Accessed November 5, 2014
6. Jochberger S, Ismailova F, Lederer W, Mayr VD, Luckner G, Wenzel V, Ulmer H, Hasibeder WR, Dünser MW‘Helfen Berührt’ Study TeamAnesthesia and its allied disciplines in the developing world: a nationwide survey of the Republic of Zambia.Anesth Analg20081069428
7. Merry AF, Cooper JB, Soyannwo O, Wilson IH, Eichhorn JHInternational standards for a safe practice of anesthesia 2010.Can J Anaesth201057102734
8. Hodges SC, Walker IA, Bösenberg ATPaediatric anaesthesia in developing countries.Anaesthesia200762suppl 12631
9. Green SM, Clem KJ, Rothrock SGKetamine safety profile in the developing world: survey of practitioners.Acad Emerg Med19963598604
10. Rukewe A, Adebayo OK, Fatiregun AAFailed obstetric spinal anesthesia in a Nigerian teaching hospital: incidence and risk factors.Anesth Analg201512113015
11. Enohumah KO, Imarengiaye COFactors associated with anaesthesia-related maternal mortality in a tertiary hospital in Nigeria.Acta Anaesthesiol Scand20065020610
12. Fenton PM, Whitty CJ, Reynolds FCaesarean section in Malawi: prospective study of early maternal and perinatal mortality.BMJ2003327587
13. Glenshaw M, Madzimbamuto FDAnaesthesia associated mortality in a district hospital in Zimbabwe: 1994 to 2001.Cent Afr J Med2005513944
14. Ouro-Bang’na Maman AF, Tomta K, Ahouangbévi S, Chobli MDeaths associated with anaesthesia in Togo, West Africa.Trop Doct2005352202
15. Li G, Warner M, Lang BH, Huang L, Sun LSEpidemiology of anesthesia-related mortality in the United States, 1999–2005.Anesthesiology200911075965
16. Thomas M, George NA, Gowri BP, George PS, Sebastian PComparative evaluation of ASA classification and ACE-27 index as morbidity scoring systems in oncosurgeries.Indian J Anaesth20105421925
17. Wolters U, Wolf T, Stützer H, Schröder TASA classification and perioperative variables as predictors of postoperative outcome.Br J Anaesth19967721722
18. Liu H, Waxman DA, Main R, Mattke SUtilization of anesthesia services during outpatient endoscopies and colonoscopies and associated spending in 2003–2009.JAMA2012307117884
19. Rosen MA, Lee BH, Sampson JB, Koka R, Chima AM, Ogbuagu OU, Marx MK, Kamara TB, Koroma M, Jackson EV JrFailure mode and effects analysis applied to the maintenance and repair of anesthetic equipment in an austere medical environment.Int J Qual Health Care20142640410
20. Rosen MA, Chima AM, Sampson JB, Jackson EV Jr, Koka R, Marx MK, Kamara TB, Ogbuagu OU, Lee BHEngaging staff to improve quality and safety in an austere medical environment: a case-control study in two Sierra Leonean hospitals.Int J Qual Health Care2015273207
Copyright © 2016 International Anesthesia Research Society