On January 12, 2010, a 7.0 ML earthquake with its epicenter in Léogâne, Haiti, approximately 16 miles west of the capitol Port-au-Prince, devastated both of these cities.a Twelve days later, we arrived at Hôpital Sacré Coeur in Milot, Haiti to care for injured patients. In the following, we offer our observations as a guide to other anesthesiologists who may wish to travel to Haiti or other countries after catastrophic events to provide anesthetic care for the injured.
Haiti is on the island of Hispaniola, which encompassed a Haitian population of approximately 10.033 million in 2009.1 The income per capita is approximately US $520/year (United States [US]: $46,040/year) with a large majority of Haitians living in poverty. The average life expectancy is 61.5 years, the lowest in the Western hemisphere (US expectancy: 79.4 years).1 The Haitian government estimates that >200,000 people died during and immediately after the January 12, 2010 earthquake and an additional several hundred thousand people were injured.b This is similar in human devastation with the Wenchuan, China 8.0 ML earthquake of 2008 that injured 370,000 people.2 Before the island earthquake, Haitian hospitals had approximately 1.3 beds/1000 population (US: 3.1 beds/1000).1 Such a large instantaneous influx of severely injured citizens would overwhelm even the most well-resourced health care system. Clearly, Haiti needed massive health care assistance.
HÔPITAL SACRÉ COEUR
We worked in Milot, a village 9 miles south of Cap Haitian and 54 miles north of Port-au-Prince, which houses the nation's largest private hospital, Hôpital Sacré Coeur (Fig. 1). Normally, Hôpital Sacré Coeur serves a local population of approximately 225,000 people with 74 inpatient beds and 2 fully functional operating rooms (ORs) allowing approximately 1200 operations in 2008. Surgeons and anesthesiologists from the US rotate through the hospital on a weekly basis to perform various operations. Locally, Haitian surgeons reside on site for routine operations (e.g., cesarean deliveries) with anesthesia provided by Haitian-trained nurse anesthetists. The typical weekday operative caseload is 5 to 7 operations/day with an emphasis on spinal anesthesia. General anesthesia is usually considered to be dangerous for a number of reasons including malfunctions of the anesthesia machine ventilator, frequent hospital generator failure, and hypoxia. Because the earthquake did not damage Milot, this disaster transformed the hospital into a major depot for trauma patients from affected areas.
In the chaotic days after the earthquake, >360 newly injured patients overwhelmed the hospital. This influx caused the hospital to become extremely disorganized. Patients scheduled for surgery were “misplaced” as new wards were opened in the adjacent schools. Patient names on posting slips were incorrect and the use of the native Creole language contributed to miscommunication. Incoming supplies from relief agencies were scattered in various places throughout the local health care system and may have been in adequate supply, but frontline providers did not know this because there was no effective supply function. The quickly recognized need for better organization prompted a multidisciplinary team approach with the designation of an anesthesiologist to lead the preoperative holding, ORs, postanesthesia care unit (PACU), and growing intensive care unit (ICU). Over several days, this organizational superstructure to the ORs and associated areas along with the interpreters allowed more systematic, efficient, and orderly care of surgical patients.
Given the pressing need for surgical care, 2 ORs were inadequate and 4 additional rooms in the OR area were converted to primitive ORs with no modern monitoring capability (Fig. 2). In these additional rooms, not even the standards of basic anesthesia monitoring could be met.3 Thirty to 35 cases/day in these locations were relegated to those that lent themselves to IV sedation and regional anesthesia. Patients in these additional rooms were assessed and monitored by traditional physical examination: chest excursion, color, carotid pulse, and manual arterial blood pressure measurement. Ten days after the earthquake, Philips Healthcare (Andover, MA) equipped, at no cost, all 6 anesthetizing locations and the PACU with modern, fully functional monitors (IntelliVue MP90 with gas modules).
Communication with patients was difficult. Creole, along with written French, is the official language of Haiti and spoken by all the native people. Written French appeared to be common in most hospital documents. Almost none of the patients and families we encountered spoke English or French. For these reasons, an interpreter was essential to perform the preoperative evaluation unlike the situation for anesthesiologists in the Wenchuan, China earthquake disaster zone.4
The medical history usually encompassed the date of injury (always January 10, 2010, the day of the earthquake), other previous medical problems (almost none of our patients had ever seen a doctor), verification of fasting status, and inquiry about drug allergies. Many patients were dehydrated or under-resuscitated, a finding also echoed by anesthesiologists caring for similarly injured patients after the Wenchuan, China earthquake.2,4 Because modern diagnostic tools were either not available or very limited, physical examination skills were the usual means for patient assessment.
Fortunately, the hospital possessed a very functional laboratory. Common blood tests including a complete blood count, electrolytes, creatinine, and glucose values could be rapidly obtained during the day. In addition, there was a small blood bank and it was supplied by local community donations, although the resident human immunodeficiency virus prevalence is estimated to be 2.0% to 3.1%.5,6 The blood bank did not fractionate their products so only whole blood was available. We did not observe adverse reactions to whole blood, a finding noted by physicians working after the Wenchuan earthquake. Point-of-care testing was performed using handheld devices brought by American medical staff, but disposable cassettes were quickly exhausted. Anemia was very common, secondary to preexisting malnutrition and severe injuries with concomitant blood loss. Radiology services consisted of 1 fixed-location machine. Radiographs took approximately 2 hours to process, the capacity was 2 per hour, and the quality was marginal. Nevertheless, these films provided adequate confirmatory evidence of many orthopedic injuries.
INTRAOPERATIVE CARE ISSUES
Timing of Medical Care
The types of injuries and medical problems encountered depended on the time from the disaster to the time that care was given. As shown in Table 1, the problems that may be seen intraoperatively differ between the acute injuries and the more chronic problems. We arrived <2 weeks after the earthquake and witnessed the transition from taking care of the fractures and other injuries to what turned out to be even more serious problems such as tetanus, diabetic ketoacidosis, and sepsis.
Expanding a Small Rural Hospital to a Trauma Center
When a disaster of this magnitude strikes a country, there is a high likelihood of having to quickly transform a small hospital into a large receiving hospital. The challenges of this task are enormous, including:
- Increasing staffing levels of physicians, nurses, aides, and interpreters (if needed) on site;
- Providing a means to communicate between triage personnel and hospital-based providers;
- Increasing the facilities and equipment needs to handle the large influx of patients;
- Procuring the large increase in supplies needed; and
- Ensuring that transportation of critically ill patients is done in the safest possible manner.
H cylinders were the only source of oxygen and were supplied by trucks from a distant location. These tanks were attached to a flow regulator for use with anesthesia machines, but there was no pressure gauge on any of the tanks. Tanks were inconsistently color coded and varied in hue: green, partially green, partially blue, brown, and white. All were rusted and most were labeled with tape reading plen (full) or vid (empty).
Medications frequently used in anesthetic practice were available in the OR suite although less frequently used drugs (e.g., flumazenil, milrinone) were not. In addition, because the manufacturing origins of the drugs were from many nations, the name and concentrations of the drugs sometimes varied from those familiar to clinicians practicing in the US and necessitated scrupulous attention to labeling. Several anesthesiologists and other physicians brought medications with them from the US, including antibiotics, oral analgesics, narcotics, local anesthetics, propofol, volatile anesthetics, ketamine, and midazolam. All medications and supplies were deposited into the OR, although oral analgesics and antibiotics were added to the hospital pharmacy stock for ward use. Because of the disorganization of supplies, both anesthesia and surgical clinical providers searched for and collected supplies from the various caches before beginning every case. We highly recommend the presence of an anesthesia technician, which would have provided a large measure of organization to ensure smoother anesthetic services.
Because of the absence of anesthesia machines and ventilators in several of our ORs, we learned that spontaneous respiration was essential for rapid emergence with the opportunity to bypass the PACU, and extremely rapid “turnover” of 5 to 10 minutes between cases. To achieve these goals with the constraints placed on us, very deep sedation was frequently employed using IV midazolam (0.05–0.10 mg/kg) and ketamine (1.0–2.0 mg/kg) with augmentation by locally administered anesthesia if possible. Occasionally, minimal bolus doses (0.3 mg/kg) of propofol were necessary as a supplement for its hypnotic effects. Similarly, Mulvey et al.7,8 reported a heavy reliance on benzodiazepines and ketamine when treating victims of the 7.6 ML Kashmir earthquake of 2005. For more complex cases requiring airway management, general anesthesia was induced with propofol with maintenance by a volatile anesthetic. For other disasters, we highly recommend a large stock of a short-acting benzodiazepine and ketamine.
Airway problems encountered included those resulting from facial trauma and burn victims. The available airway equipment was limited because general anesthesia was infrequently performed before the earthquake in the hospital. We encourage others who enter a disaster area such as this to bring their own airway management devices with which they are most facile as backup airway equipment.
Many buildings in the Port-au-Prince area were made of unreinforced masonry, which is similar to Wenchuan.9 Because of these structural problems, many of the injuries were extremity fractures amenable to regional anesthesia, a view also reinforced by anesthesiologists caring for Chinese patients after the 2008 earthquake.2,4 In addition, minimal to no postoperative IV analgesics were available for patients with moderate to severe pain. To a large extent, lower extremity procedures were ideally suited to spinal anesthetics. In addition, some of these patients were discharged from the OR directly to the wards, especially as the PACU evolved to become an ICU for the entire hospital. Although some spinal anesthesia kits were available, the most cost-effective and smallest logistical “footprint” for travel were spinal needles and ampules of local anesthetic.
We performed infraclavicular brachial, femoral nerve, popliteal nerve, saphenous nerve, and ankle blocks with a nerve stimulator because no ultrasound was available. Many patients with extremity injuries would have benefited from the use of perineural sheath catheters with continuous infusion of local anesthetic after orthopedic surgery.10 We did not have the requisite equipment to place these catheters nor were the ward nurses familiar with the postoperative care. In addition, we contemplated making “pain rounds” in the evenings, but this idea failed given the logistical and communication obstacles. Regional anesthesia was extremely useful to provide operative anesthesia and postoperative analgesia while reducing the burden on the PACU.
Postoperative care was available in an 8-bed recovery room. As with the ORs, Philips Healthcare equipped this location with modern monitors. Auxiliary oxygen required that an H cylinder of oxygen be placed in the PACU. Many patients could use a single oxygen cylinder if multiple valves were “daisy-chained” from the pressure regulator. Some fatalities occurred in the PACU/ICU during our tenure. These included individual patients with fulminant tetanus, diabetic ketoacidosis, pulmonary aspiration, renal failure after rhabdomyolysis, and many cases of sepsis. There were simultaneously 3 patients with hemoglobin concentrations <3.0 mg/dL in the ICU. One patient had a seizure that we interpreted to be an anemic seizure, an event none had witnessed previously and has not otherwise been reported. The presence of a critical care medicine physician would have been a welcome addition.
The primary mode of transportation for injured patients coming from the quake area was the US Navy. The Seahawk (the US Navy version of the US Army's Blackhawk) helicopter made flights from the earthquake area to our hospital numerous times a day. The local soccer field became the focal point for initial evaluation of injured patients. Orthopedic surgeons would assist with initial evaluation, because the vast majority of our patients had long bone fractures. These fractures were often open and infected, a situation previously reported with the 2005 earthquake in Kashmir, with a 78% incidence of limb injury.7,8 We observed few chest or abdominal injuries likely because such patients did not survive the interval before we arrived. With no ability to ventilate patients' lungs long-term and extremely limited intensive care capability, we attempted to transfer patients to the only facility capable of caring for these people, the USS Comfort. That ship was extremely busy and the demand for care usually exceeded its capacity. We attempted to transfer an intubated child (continuously hand-ventilated because of absence of a ventilator) to the USS Comfort but the child died during transport.
Professionalism endures as a hallmark of the practice of anesthesiology. Four specific aspects of this trait emerged during our service in Haiti. First, compassion and sensitivity to the plight of injured Haitian patients and their families was necessary. From the family's point of view, an injured loved one may have been trapped under rubble in Port-au-Prince, been evacuated to Milot by helicopter, and admitted to a new hospital without their knowledge. Second, the foreign medical staff was sensitive to Haitian health care practitioners. Haitian medical staff operated the hospital where we worked. We were visitors in their hospital and remained cognizant of our status as guests while concurrently providing the care. To that end, we tried to cooperate in maintaining accurate OR logs, use proper pharmacy and laboratory requests, and abide by other hospital policies. Third, we needed the ability to adapt an anesthetic plan to changes in patient status and availability of equipment and supplies. This flexibility was evident for anesthesia providers who repaired anesthesia machines, reappointed oxygen cylinders for multiple patients in the PACU/ICU, and provided emergent anesthesia in some cases without monitors or oxygen. Finally, medical staff met nightly to discuss daily challenges and possible solutions. This feedback was important to follow-up planning and led to immediate changes in operations.
During our stay at Hôpital Sacré Coeur, we felt perfectly safe. Even at night, the women in our group felt safe enough to walk the quarter mile from the residential compound where we lived to the hospital. However, in the course of our week, there was a noticeable increase in the people living in the streets; most of these people had moved from the earthquake zone. The future security environment is unknown and dependent on the United Nations troops, Haitian military and police, and nongovernmental organizations. Security concerns are not unique to Haiti. After the earthquake in Wenchuan, the Chinese government quickly moved troops into the disaster area.11 With mounting desperation for food and shelter, it is unknown how secure these health care facilities will be in the coming months.
Preparation and Survival
Tropical diseases are a fact of life in Haiti, but receiving the appropriate immunizations and arranging for prophylaxis before traveling was straightforward. If hepatitis B and tetanus immunizations were current, additional prophylaxis was required for hepatitis A and typhoid. Because malaria is common in Haiti, our group members took doxycycline. Tuberculosis is endemic in Haiti. Although there is no prophylaxis, we were advised to take N95 masks. Finally, with approximately 2.0% to 3.5% of the Haitian population positive for human immunodeficiency virus, universal precautions were necessary to avoid contact with body fluids.5,6 The most important resource in a natural disaster such as this is water.12 We were fortunate that the United Nations kept us well supplied because the local drinking water was not safe. Before entering a disaster area, make sure your supply of water is adequate.
Electricity at the hospital was supplied by diesel generators. There are no landlines and no power grids in this area. At the hospital, intramedical staff communications were limited to 20 2-way radios. Cell phones worked intermittently and were highly position-dependent as was the compound's only satellite phone. Internet using a satellite high-speed system provided 5 Ethernet stations attached to a variety of computers. The Internet became a very important source of information, with searches ranging from treating an unfamiliar disease such as tetanus to determining the volume of an H cylinder.
CONTRAST WITH THE WENCHUAN EARTHQUAKE
Although the injuries we encountered were similar to those described with the Wenchuan earthquake, there are a number of important differences with responding to a natural disaster in a third-world foreign country, as depicted in Table 2, compared with responding to a calamity in your own country.
In summary, the Haitian people needed massive health care assistance of every kind. We were fortunate to have the opportunity to provide anesthetic care in an undamaged hospital to a large number of extremely grateful patients suffering from mostly orthopedic injuries. Although challenging from many perspectives, the experience was emotionally enriching and recalls the fundamental reasons why we selected medicine and anesthesiology as a profession. We hope that others will likewise embrace this fulfilling opportunity and serve.
The authors thank Nikolaus Gravenstein, MD, for editorial comments and our colleagues whose flexible scheduling allowed us to work in Haiti.
a Romero S, Lacey M. Fierce quake devastates Haitian capital. The New York Times, January 12, 2010.
b Baker P, Berger J. US to resume airlift of injured Haitians. The New York Times, January 31, 2010.
1. Pan America Health Organization. Health Situation in the Americas: Basic Indicators 2009. Washington, DC: World Health Organization, 2009
2. Jiang J, Xu H, Liu H, Yuan H, Wang C, Ye J. Anaesthetic management under field conditions after the 12 May 2008 earthquake in Wenchuan, China. Injury 2010;41:e1–3
3. ASA House of Delegates. Standards for Basic Anesthetic Monitoring. American Society of Anesthesiologists, Park Ridge, IL. 2005. http://www.asahq.org
4. Chen G, Lai W, Liu F, Mao Q, Tu F, Wen J, Xiao H, Zhang JC, Zhu T, Chen B, Hu ZY, Li RM, Liang Z, Nie H, Yan H, Yang BX, Du Q, Huang WX, Jiang YW, Kwan AS, Song L, Wu CM, Xiang T, Xu HW, Lau WB, Song HB, Wen CB, Yao ZH, Zhang L, Zeng J, Dai YE, Lopez BL, Zheng JQ, Zhou J, Christopher TA, Ma XL, Yu H, Xu LL, Guo Q, Song ZP, Volinn E, Kryger K, Cao Y, Ge H, Liu H, Luo CZ, Tao W, Zuo YX, Liu J. The dragon strikes: lessons from the Wenchuan earthquake. Anesth Analg 2010;110:908–15
5. Halperin DT, de Moya EA, Perez-Then E, Pappas G, Garcia Calleja JM. Understanding the HIV epidemic in the Dominican Republic: a prevention success story in the Caribbean? J Acquir Immune Defic Syndr 2009;51:S52–9
6. Figueroa JP. The HIV epidemic in the Caribbean: meeting the challenges of achieving universal access to prevention, treatment and care. West Indian Med J 2008;57:195–203
7. Mulvey JM, Qadri AA, Maqsood MA. Earthquake injuries and the use of ketamine for surgical procedures: the Kashmir experience. Anaesth Intensive Care 2006;34:489–94
8. Mulvey JM, Awan SU, Qadri AA, Maqsood MA. Profile of injuries arising from the 2005 Kashmir earthquake: the first 72 h. Injury 2008;39:554–60
9. Hudnut KW. ShakeOut, California style. Anesth Analg 2010;110:655–6
10. Ilfeld BM, Morey TE, Enneking FK. Continuous infraclavicular perineural infusion with clonidine and ropivacaine compared with ropivacaine alone: a randomized, double-blinded, controlled study. Anesth Analg 2003;97:706–12
11. Guidry OF. Deja vu. Anesth Analg 2010;110:651–2
12. Lynch DK. Plan ahead. Anesth Analg 2010;110:653–4
Funding and supplies for this activity were provided by Shands Hospital at the University of Florida (Gainesville, FL), the CRUDEM Foundation (Lowell, MA), and unknown aircraft owners and operators who donated time and fuel.
All authors helped write the manuscript and all authors approved the final manuscript.