Abstract
The aim of this study was to describe the frequency and features of headache among patients with confirmed dengue virus infection and to compare the headache features in patients with dengue fever and dengue haemorrhagic fever, primary and secondary dengue infection, and patients with and without neurological involvement. Patients with classic dengue fever had a more intense headache than those with the more severe form of the disease, dengue haemorrhagic fever.
Introduction
Dengue is the most frequent human arboviral infection, with nearly 50 million infections occurring annually. The disease is caused by a flavivirus transmitted in urban areas by the female of Aedes aegypti (1). This mosquito breeds in domestic collections of clean water (2). There are four distinct dengue virus serotypes. A primary infection by one of them confers lifelong immunity against that serotype, but secondary infection by one of the other serotypes may still occur. The risk of a serious disease increases if it is a secondary infection. This occurs because the pre-existing non-neutralizing antibodies against a previous serotype may increase the number of infected monocytes, resulting in more cell-presented dengue viral antigens to T lymphocytes and a more intense activation of the immune response (1–4).
Clinically, dengue virus infection ranges from an asymptomatic infection to severe haemorrhagic disease with shock. Classic dengue fever (DF) generally presents with high fever, severe headache, retro-orbital pain, myalgia, arthralgia, nausea and rash. Patients with dengue haemorrhagic fever (DHF) have a more severe evolution with haemorrhagic manifestations, hepatomegaly, haemoconcentration, severe thrombocytopenia and shock. Neurological manifestations are rare and are found more frequently in patients with DF. Encephalopathy with confusion, drowsiness and seizures are the most frequent findings. There are also true encephalitis cases in which the virus can be identified post mortem in brain tissue or during the course of the disease and in which the viral RNA can be identified in cerebrospinal fluid (CSF) (5–10).
Headache in patients with dengue has been described as a severe, frontal and retro-ocular pain (1, 10). In previous studies headache was found in more than 95% of patients with dengue infection. It has been shown that headache and retro-ocular pain are as frequent in DF as in DHF (11). However, there are no previous reports comparing headache intensity and features in patients with DF with those in patients with DHF. Also, it is unknown if headache features in patients with DF and patients with primary or secondary dengue infection. The aim of this study was to describe the frequency and features of headache in a group of Brazilian patients with confirmed dengue infection, as well as to compare headache intensity and features among patients with DF and DHF, primary and secondary infection, and patients with and without neurological findings.
Patients and methods
All patients with suspected dengue infection admitted to the Santa Casa de Misericórdia de Vitória between October 2002 and February 2003 were submitted to serum sample collection and a standardized clinical evaluation protocol. The diagnosis of dengue was established by at least one of the following criteria: a positive serum IgM or a fourfold rise in serum IgG titre (12). Serum reverse transcriptase-polymerase chain reaction (RT-PCR) was performed in order to determine the dengue virus serotype. Clinical evaluation included history, physical examination, neurological examination and a headache questionnaire applied to all patients. Patients were classified as having primary or secondary infection according to the IgG affinity test (patients with primary infection have lower affinity anti-dengue IgG antibodies) (13). Diagnoses of DF and DHF were established according to World Health Organization criteria (14). Patients with altered mental status, seizures or confusion were considered to have encephalopathy. When encephalopathy was present and viral RNA was identified in a CSF sample a diagnosis of encephalitis was made (15).
Patients were asked about headache localization, intensity, time of duration and associated symptoms (nausea, vomiting, photophobia and phonophobia). The intensity of headache was recorded according to the following scale: 0, absent; 1, mild (not disabling); 2, moderate (mildly disabling); 3, severe (considerably disabling); and 4, extremely severe (completely disabling). Headache intensity and features were compared between the group of patients with primary infection and that with secondary dengue virus infection. The intensity and features of headache were compared between patients with and those without neurological involvement. Also, headache intensity and features were compared between patients with DF and patients with DHF. All statistical comparisons were performed with Fisher's exact two-tailed test. The study was approved by the Ethics Commission of Santa Casa de Misericórdia. Written informed consent was obtained from all participating subjects.
Results
The diagnosis of dengue was confirmed in 83 cases, by positive IgM in 30, fourfould rise in IgG titre in 14 and by both in 49. The mean age was 42 ± 23.9 years and most patients (61.7%) were female. Eighty-one (97.6%) of the patients with confirmed dengue reported to have headache. All of the patients (100%) had fever. The mean duration of febrile disease was 6 ± 2.6 days and that of headache was 5 ± 2 days. PCR was performed in 82 patients and was positive in 44. Dengue virus type 3 was the most frequently seen (97.7%). It was identified in 42 cases, type 1 in one, with one patient coinfected with types 1 and 3. The neurological evaluation was considered abnormal in 18 patients. Sixteen patients had encephalopathy, one had meningismus and one had peripheral neuropathy. Only 13 of these patients were submitted to a spinal tap. Only the patient with meningismus had an increased CSF leucocyte count (130 cells/mm3); the others had normal CSF cell, glucose and protein evaluations. CSF PCR for dengue virus was positive in six out of the 13 CSF samples, five with encephalitis and one with peripheral neuropathy. Dengue type 3 was found in five and type 1 in one of them.
Fifty-three (66.25%) patients had primary dengue virus infection and 27 (33.75%) were considered to have secondary infection. Only 18 (21.7%) of the patients were diagnosed as having DHF and 65 (78.3%) had classic dengue fever.
Patients with headache presented with pain most frequently in the frontal region (65.4%), followed by retro-ocular (49.4%), diffuse (29.6%), occipital (19.7%), neck (13.6%) and temporal (8.6%) regions. In all patients headache was bilateral. Forty-eight patients (59.2%) reported a throbbing headache and 33 (40.7%) a pressing/tightening headache pattern. Pain was considered mild by one patient (1.2%), moderate by 16 (19.7%), severe by 22 (27.1%) and extremely severe by 42 (51.8%). Nausea and/or vomiting were found in 70 (86.4%) of the patients. Photophobia was found in 45 patients (55.5%) and phonophobia was mentioned by 51 (62.9%). The headache was aggravated by exposure to light in 19, odours in 11, physical activity and Valsava in eight. Forty-five patients did not report aggravating factors for the headache. All patients had complete remission of the dengue symptoms including headache. Fourteen patients described an improvement with dypirone and 33 with acetaminophen, but 36 declared that analgesics were inefficient. The most frequent clinical manifestations of these patients were: muscle pain (98.8%), malaise (97.5%), joint pain (38.3%), diarrhoea (22.2%), abdominal pain (24.7%), petechiae and purpura (4.8%), rash (20.7%), gingival bleeding (19.7%) and gastrointestinal tract bleeding (12.3%).
There were no significant differences in the headache intensity and features between patients with primary or secondary dengue infection (P > 0.05). Headache intensity and features were not significantly different between patients with and patients without neurological involvement (P > 0.05). Headache features were not significantly different between patients with DF and those with DHF (P > 0.05). However, patients with classic dengue fever had a more intense headache than patients with DHF (P < 0.05).
Discussion
Dengue has been an important epidemic disease in our region over the last decade. It has had severe social and medical impact with hundreds of hospitalizations and some deaths. Dengue virus type 3 was the most prevalent serotype during the 2002–2003 epidemics. In our study we have found headache in almost all patients with dengue and it was characterized by a very intense, bilateral, throbbing, frontal and/or retro-orbital pain. Most patients had associated symptoms such as nausea and/or vomiting, photophobia and phonophobia. Nausea and/or vomiting, photo- and phonophobia are not necessarily related to the headache since the infection could itself cause these symptoms. However, these are routinely investigated symptoms in headache questionnaires and the high frequency of these associated symptoms may be important in the differential diagnosis of dengue-associated headache from other primary or secondary headaches. The dengue headache has some features in common with migraine, because it is most commonly throbbing and is usually associated with nausea, photophobia and phonophobia (16). However, the presence of associated clinical findings such as fever and muscle pain allows an easy distinction between these two conditions. In epidemic areas for dengue virus infection where dengue and migraine are prevalent, it is important to investigate carefully for the presence of fever, muscle pain and other dengue fever signs. This distinction is especially relevant because some migraine patients take acetylsalicylic acid for headache relief, but patients with active dengue infection can not take this drug due to an increased risk of haemorrhagic manifestations. Treatment of dengue headache is with common analgesics, such as acetaminofen. Fluids replacement, fever control, blood pressure and level of consciousness are also very important (1).
The IHS considers two different headaches related to viral infections. The first headache is attributed to lymphocytic meningitis (9.1.2) and to encephalitis (9.1.3) and is due to the inflammatory reaction to the virus in the central nervous system or meninges. The second is the headache attributed to systemic viral infection (9.2.2). This last one may be a consequence of the fever or of systemic production of substances that may activate central nervous system mechanisms that determine headache, e.g. inducible nitric oxide synthase resulting in nitric oxide production (17). Headache in dengue can not be attributed to lymphocytic meningitis because meningitis was found in just one patient. Encephalitis is not the headache explanation, since only 16 of our patients had encephalopathy and only five of them had positive CSF PCR and could be classified as having true encephalitis. This confirms previous reports that headache is due to systemic infection and not due to direct viral invasion of brain parenchyma and meninges (17).
It has been suggested that secondary infection with a different serotype of dengue virus is associated with a more severe disease (1), due to a more intense immune reaction in these patients. Since headache results from systemic infection and inflammation it would be reasonable to expect a more intense headache in patients with secondary infection. However, our results show that headache intensity and headache features were not different in patients with primary and secondary infection. In DHF there is an increased vascular permeability, leading to plasma leakage to the tissue. There are also haemorrhagic manifestations. These pathophysiological features are due to complex events with intense production of interferon-γ, tumour necrosis factor-α and complement activation on endothelial cells (1). If this intense inflammation can influence headache severity and features it should be expected that patients with DHF will have a more severe headache than patients with classic dengue fever. However, headache features were not different between patients with DF and DHF and headache was unexpectedly more severe among patients with DF. Many authors have postulated that the pathogenesis of neurological symptoms of DHF and DF is different. In DHF the symptoms are related to liver failure and oedema through the cerebral vasculature. In classic DF the pathogenesis of neurological involvement is less clear. Immunopathological mechanisms or neurotropism may be determinant factors in different patients (17). It is possible that these differences are also responsible for the more intense headache among patients with classic DF.
There are limitations to our study. We did not assess the previous history of primary headache in our patients. Future studies should evaluate if dengue headache in patients with previous primary headache such as migraine is different from dengue headache in those without. We were not able to perform imaging studies, but they might have been helpful in providing information about the mechanism of central nervous system involvement in dengue as well as the headache pathogenesis in this infection.
In conclusion, headache is one of the most frequent and disabling symptoms in dengue infection. Headache intensity was not related to the severity of the disease but headache was more intense in patients with classic DF than in those with DHF. Since most of our patients were infected with dengue virus type 3, future studies are needed to determine the headache features in infections with the other dengue virus serotypes.