4. DIAGNOSIS
4.1 Clinical diagnosis of VHF
Signs and symptoms of VHF
Early signs and symptoms are non-specific, and patients may present with fever, headache, conjuctivitis, pharyngitis, myalgia (especially lower back pain), vomiting, abdominal pain and diarrhoea. Recognition of the syndrome is easier once patients develop a petechial rash or ecchymoses, and other haemorrhagic signs such as epistaxis, haematemesis and melaena. There may be rapid progression to multi-organ failure, altered mental state, jaundice and shock.
Important information to bear in mind during clinical diagnosis
Not all patients with VHF bleed, and it is more important to recognize a syndrome that may include bleeding, nosocomial transmission, evidence of thrombocytopenia and hepatic dysfunction, notably raised transaminases.
Clinicians can seek advice from the medical officer on duty at the National Institute for Communicable Diseases (NICD) (cellular telephone number 082 883 9920).
More than 90% of suspected cases of VHF prove to be severe forms of common diseases. Many of the diseases mistaken for VHF are treatable if diagnosed early. There must be systematic elimination of differential diagnoses (see section 4.2).
Failure to institute appropriate safety precautions can have severe consequences. However, the unnecessary institution of isolation precautions is expensive and highly disruptive.
By the time that VHF is suspected patients have often received prior medical attention during which certain clinical pathology and microbiological tests may have been performed (see section 4.2).
Obtaining a history of possible exposure to infection can be crucial to diagnosing VHF. Relatives and cohorts often provide more reliable information than severely ill patients.
Detailed and accurate information required during diagnosis
● Age, sex, and place of residence of the patient (VHF infection has not yet been confirmed to have occurred within South Africa in a child <10 years old).
● Chronic medical conditions and medication, including recent drug and dosage adjustments.
● History of the current illness, including results of prior medical and laboratory investigations.
● Occupation of the patient and possible exposure to infection as in:
§ Health care and laboratory workers who tended, or processed specimens from, patients with confirmed or suspected VHF or undiagnosed fever compatible with VHF; and
§ Contact with animals or animal tissues by abattoir workers, veterinarians, farm workers, hunters, taxidermists, or persons who work with hides and skins.
● Non-occupational contact with known or suspected cases of VHF, or undiagnosed fever.
● Non-occupational contact with animals or their tissues including blood.
● Residence in or recent travel to tropical or rural environments.
● Handling or being bitten by ticks or insects, especially mosquitoes.
● Recent travel to a country known or likely to be endemic for VHF, particularly involving rural environments and contact with animals or insects - but remember that some rodent- associated and mosquito-borne VHF viruses can occur in urban environments (see section 3).
● Record exact details of:
§ The date/s of potential exposure/s to infection.
§ The date of onset of illness (incubation periods are <1 week for arbovirus infections including Congo fever, but up to 3 weeks for arenavirus, hantavirus and filovirus infections - see section 3).
§ The dates and types of all specimens previously taken and submitted for laboratory examination.
§ The results of all clinical pathology and microbiological tests already performed (see section 4.2).
Features that support a diagnosis of VHF
● Short duration and rapid progression of the disease: i.e. acute rather than chronic illness.
● Lack of evidence in the patient's history or physical examination, which excludes VHF.
● Laboratory evidence of leucopenia, thrombocytopenia, coagulation abnormalities, and raised serum transaminases, but leucocytosis can occur in CCHF, Lassa, Marburg and Ebola haemorrhagic fevers, and relatively normal platelet counts can be seen in Lassa fever.
· The progression of the illness and the timing of bleeding in relation to the onset of symptoms may be important in guiding the diagnosis of VHF versus alternative diagnosis. For example: patients with Congo fever typically bleed three to five days after the onset of illness while patients with meningococcal disease typically bleed within 24 hours after the onset of symptoms.
Features which tend to exclude a diagnosis of VHF
Normal platelet counts and normal serum transaminase levels render VHF unlikely. Confirmation of an alternative diagnosis, e.g. a positive blood culture may also render VHF unlikely. However, it is important to remember that bacterial septicaemia can occur as a complication to VHF, and in areas where malaria is endemic patients may test positive for malaria on blood smears while suffering from other infections, including VHF.
A scoring system found to be useful in the diagnosis of Congo fever in South Africa is presented in Table 4.1, and a document on answers to frequently asked questions about the disease, compiled for people involved in the livestock industry, is included as Appendix 1.
The outcome of the initial assessment may be inconclusive, but the aim should be to decide whether or not to proceed on the assumption that VHF may be involved. The disruptions and expense caused by false alarms should be balanced against the potentially dire consequences of failure to recognize VHF.
For submission of specimens for specific laboratory confirmation of VHF see section 4.3.
4.2 Differential diagnosis of suspected VHF
Procedure to follow when VHF is suspected
When VHF is suspected, it is important to obtain and interpret the results of all medical examinations and laboratory tests already performed, but warn laboratory personnel of the suspected diagnosis and ensure that further laboratory tests are only performed with appropriate biosafety precautions (see section 6.2). Another crucial step to take is to ensure that all specimens previously submitted to laboratories are retained for onward transmission to NICD along with newly collected specimens for specific VHF diagnostic tests (see.section
4.3).
Diseases commonly confused with VHF
Malaria, trypanosomiasis, relapsing fever, plague, yellow fever, other arbovirus infections and leptospirosis, especially after travel to or residence in rural or tropical areas (malaria is most common and can be rapidly fatal if not treated, but it also occurs together with other infections including VHF).
Bacterial septicaemias resemble VHF and can be rapidly fatal if not treated; most commonly caused by meningococci, but also by a wide variety of Gram-positive and - negative bacteria, and include typhoid, anthrax, and Capnocytophaga species (dysgonic fermenter 2) infection after dog bite, (septic abortion and tuberculosis with haemoptysis can also resemble VHF).
Rickettsioses: tick bite fever (TBF), Q fever, typhus; TBF often occurs in town dwellers who visit rural environments, but can also result from exposure to kennel ticks in urban settings, even where dogs are kept indoors in apartment buildings; TBF can run a fatal course very similar to Congo fever, but has an incubation period of 7-10 days after tick bite as compared to 1-3 days for Congo fever, there is usually a necrotic eschar at the site of the tick bite in TBF and the petechial rash extends to palms and soles; TBF can be treated with broad-spectrum antibiotics.
Hepatitis A, B, E, and less often C (westerners travelling in Africa often develop hepatitis A).
Fulminant systemic herpes simplex virus infection with hepatitis (with/without vesicular rash); about 60 cases have been seen in RSA with high fatality, mostly in ostensibly healthy young adults; extremely high transaminase levels which may fall terminally after virtually complete destruction of hepatocytes. Less common are severe cytomegalovirus, E-B virus or varicella-zoster virus infections, or haemorrhagic measles.
HIV seroconversion sickness, or HIV/AIDS with secondary infections, especially septicaemias.
Drug sensitivities and overdoses including anticoagulants (warfarin), other poisons and toxins including haematoxic snake bite envenomation (e.g. boomslang), industrial and agricultural chemical poisoning.
Malignant disease, e.g. leukaemia, lymphoma. Idiopathic thrombocytopenic purpura.
Heat stroke.
Interpretation of clinical pathology results for differentiating VHFs from other diseases
Full haematological examination: Findings compatible with VHF include leucopenia, thrombocytopenia, anaemia, altered clotting parameters and increased fibrin degradation products or D-dimers, but disseminated intravascular coagulopathy also occurs in many other conditions, including septicaemia. Granulocytosis suggests bacterial infection, but leucocytosis can occur in CCHF, Lassa, Marburg and Ebola haemorrhagic fevers (see section 3), and in leukaemia.
Examination of a stained blood smear: Malaria, trypanosomiasis, other haemoparasitic diseases and certain bacterial septicaemias (meningococcus, Capnocytophaga, anthrax) can be diagnosed, and differential white cell counts can be performed to provide an indication of leucocytosis/granulocytosis, leucopenia, leukaemia, anaemia, and even thrombocytopenia.
Bacteriological blood cultures: It is important that blood cultures should be performed to exclude septicaemia. Samples should be taken before antibiotic therapy is instituted. Septicaemia can be secondary to many conditions including pneumonia, gastroenteritis, perforated ulcers, and abscesses or wound infections.
Clinical chemistry tests: Raised serum transaminase levels occur commonly in VHF, and to a lesser extent also raised bilirubin levels, but jaundice and hepatocellular damage have many causes. Extremely high transaminase and bilirubin levels occur in systemic herpes simplex infection with hepatitis. Evidence of severe liver damage is a poor prognostic sign. Proteinuria is common in VHFs, notably in Lassa fever.
Specific serodiagnostic tests for non-VHF diseases: Serological tests results should be interpreted with caution, taking into account the sensitivity and specificity of the test and the stage that they are performed during the course of the illness. Notably negative results using the currently available tests for tick bite fever may not exclude the disease. Anti-HA IgM, HbsAg, HBeAg and Anti-HBc are important screening tests for hepatitis A and B. Serodiagnostic tests are available for leptospirosis, salmonellosis, measles, herpesvirus infections and many other diseases which could be confused with VHF. Rapid serum latex agglutination tests can be used to detect bacterial antigen in meningococcal septicaemia.
More than one pathology may be present in a patient, and epidemiological information and clinical laboratory findings should guide the diagnostic process.
4.3 Laboratory verification of VHF
Specific diagnostic tests for the formidable (Class 4) VHFs are performed only by the Special Pathogens Unit (SPU) at NICD. It is essential that arrangements are made directly with one of the SPU laboratory diagnosticians before specimens are submitted (Laboratory telephone numbers 011 386 6339, 082 903 9131, 082 908 8042 and 082 908 8046; NICD Hotline 082 883 9920), particularly where urgent investigations are warranted after normal work hours (07h30-16h00 Monday to Friday). The staff must be informed of the means of transport of the specimens, tracking or waybill numbers, and expected date and time of delivery.
4.3.1 Source and nature of specimens: Clinical laboratories
All specimens that may have been submitted to haematology, microbiology, clinical chemistry and other laboratories before VHF was suspected must be traced and redirected to NICD for virological examination. These specimens are important because VHF viruses are often only present in blood and other tissues in the early stages of the disease, and may be absent later.
Live patients
Specimens to be taken from live patients specifically for the investigation of suspected VHF should include 5-10ml of clotted blood and 5ml of blood taken with EDTA/sequestrene (lavender top). Throat swabs in viral transport medium may also be useful. Daily samples collected from patients in whom a diagnosis of VHF has already been confirmed provide valuable information, but need not be submitted for urgent tests; the samples can be kept refrigerated and sent to NICD in batches by routine laboratory delivery services with appropriate packaging (see 4.3.2 below).
Corpses
There is usually reluctance to proceed with a full autopsy until VHF can be excluded, and there is a widespread misconception that post mortem procedures may only be performed with the consent of relatives. However, in terms of the Health Act 61 of 2003 autopsy and removal of organs or tissues 'for determining the cause of death’ may be authorized by the medical practitioner in charge of clinical services in the hospital or authorized institution, or of the mortuary, or by a medical practitioner authorized by the person in charge of such hospital or authorized institution. Minimal specimens taken to eliminate VHF should include blood collected by cardiac puncture and liver samples taken with a biopsy needle; some liver should be placed in fixative for histopathological examination and some placed in a small volume of viral transport medium or physiological saline for virological examination. If possible, some liver tissue should also be placed in 2.5% glutaraldehyde fixative for electromicroscopy. The specimens can be taken in the ward where the death occurred or in a mortuary. Blood tends to ooze from needle puncture sites and these should be taped or sealed (e.g. Opsite®, S & N Pharmaceuticals Pty Ltd). The body should be decontaminated and sealed in double stout plastic body bags as discussed in section 6.5.
Labels attached directly to the primary specimen containers (e.g. blood tubes) should be marked clearly with the name of the patient and date of collection of the sample. For removal from the patient facility or mortuary, the specimens should be double-wrapped in zip-lock specimen bags or ordinary clear plastic bags and labeled appropriately, preferably with biohazard stickers to alert staff to the contents, and should be delivered by hand directly to the laboratory responsible for forwarding the specimens to NICD.
It may be useful to have a histopathologist examine rapidly fixed (heated formalin) and sectioned liver specimens. Bacterial septicaemia can sometimes be recognized and differentiated from liver disease due to VHF or other causes. Lack of liver lesions suggests that VHF is not involved.
4.3.2 Packaging of specimens for transfer to NICD
UN/WHO approved shipping containers for hazardous specimens are commercially available, e.g. SAF-T-PAK®, or else safe packaging can be improvised as indicated in the text box below (Figures 4.1; 4.2):
The method used for transmitting specimens to NICD depends on the urgency with which diagnostic tests are required, proximity to NICD, and the availability and speed of routine delivery services for transmitting specimens to NICD as operated by the National Health Laboratory Service (NHLS) and private companies (e.g. Ampath, Lancet)
For the delivery of specimens for urgent tests from within a few hours distance by road from NICD, it may be necessary to assign a specific vehicle and driver. This applies even to hospitals within close proximity to NICD since routine specimen delivery routes are operated at certain times of day only. Sometimes relatives of patients are willing to deliver specimens when no other rapid means of transport is available. Specimens should be delivered directly to members of SPU staff (contact telephone numbers: 011 386 6339, 082 903 9131, 082 908 8042, 082 908 8046; NICD Hotline 082 883 9920), or after hours left with the security guards at the entrance to NICD by prior arrangement with SPU staff (for map see Figure 4.6).
For delivery of specimens from longer distances it may be possible to utilize routine laboratory delivery services, or a commercial courier service using scheduled road or air transport and door-to-door delivery, depending on the urgency with which tests are required. However, deliveries after normal work hours, and particularly at weekends, can be difficult to arrange.
Follow up specimens from patients in whom the diagnosis has already been confirmed or sera from healthy contacts of VHF patients which are sent for routine screening and do not require urgent tests, can be sent to NICD by regular laboratory delivery services with appropriate packaging.
4.3.3 Laboratory tests
If emergency tests are warranted and appropriate arrangements have been made ahead of time with SPU staff (telephone numbers 011 386 6339, 082 903 9131, 082 908 8042 and 082 908 8046; NICD Hotline 082 883 9920) tests can be performed after normal work hours, which are 07h30-16h00 on weekdays only.
4.3.4 Interpretation of results
In the acute phase of the disease, cases of VHF are diagnosed by identifying virus antigen or nucleic acid in the specimens, or by isolating (culturing) live virus. Virus antigen detection tests are used for certain diseases only and take 3-8 hours to complete. Detection of virus nucleic acid by reverse transcription-polymerase chain reaction (RT-PCR) takes 6-12 hours from the time of receiving the specimen in the laboratory, depending on whether or not there is need for nested (second round) tests. Isolating virus in culture can sometimes be achieved within 2 days but usually takes a week or longer.
In the convalescent phase of the disease, cases of VHF are diagnosed by identifying an antibody response. Preliminary IgG antibody tests can be completed within two hours of receipt of specimens and IgM tests within 3 hours, but overnight tests produce more reliable results.
All serum samples (acute and convalescent) are routinely tested for antibodies to the full range of African VHF viruses. This is because the clinical histories received are sometimes inaccurate, particularly with respect to the date of onset or duration of illness.
It is extremely important to remember that even acute specimens for which virus antigen, RT-PCR and antibody tests are all negative, occasionally yield virus in culture some days later. Failure to appreciate this possibility has led to serious misunderstandings in the past.
Sometimes it is necessary to submit a further sample to clarify an ambiguous finding. For example, detection of IgG antibody on its own, without virus or IgM antibody, could indicate past infection not connected to the current illness, but sometimes IgG can appear in circulation slightly before IgM during convalescence.
It is almost equally important to eliminate a possible diagnosis of VHF as it is to confirm a diagnosis rapidly: failure to detect virus or viral nucleic acid in serum during the first 7 days of illness, or to demonstrate antibody two weeks after onset, constitutes a fair indication that one of the known African VHFs is not involved. However, viraemia may be of very short duration or absent. Hence, negative findings on samples taken early in the course of disease should be supported by antibody tests on further specimens taken in convalescence.
In emergencies results are made known telephonically or by fax as soon as possible, with written confirmation following later (remember to include contact details for the person to whom results should be reported when submitting specimens).
