Corticosteroides para la meningitis eosinofílica parasitaria
Resumen
Antecedentes
Angiostrongylus cantonensis (A.cantonensis) es la causa principal de la meningitis eosinofílica infecciosa. Las larvas muertas de este parásito provocan inflamación y exacerban los síntomas de la meningitis. Los corticosteroides son fármacos utilizados para reducir la inflamación causada por este parásito.
Objetivos
Evaluar la eficacia y la seguridad de los corticosteroides en el tratamiento de la meningitis eosinofílica.
Métodos de búsqueda
Se hicieron búsquedas en CENTRAL (2014, número 11), MEDLINE (1950 hasta la tercera semana de noviembre de 2014), EMBASE (1974 hasta diciembre de 2014), Scopus (1960 hasta diciembre de 2014), Web of Science (1955 hasta diciembre de 2014), LILACS (1982 hasta diciembre de 2014) y CINAHL (1981 hasta diciembre de 2014).
Criterios de selección
Ensayos controlados aleatorizados (ECA) de corticosteroides versus placebo para la meningitis eosinofílica.
Obtención y análisis de los datos
Dos autores de la revisión (SiT, SaT), de forma independiente, recopilaron y extrajeron los datos de los estudios. Se calificó la calidad metodológica de los ECA. Se identificaron y analizaron los resultados y los efectos adversos.
Resultados principales
No se identificaron nuevos ensayos para su inclusión en esta actualización de 2014. Un estudio que incluyó a 110 participantes (55 participantes en cada grupo) cumplió los criterios de inclusión. El corticosteroide (prednisolona) mostró un efecto beneficioso al acortar la mediana del tiempo hasta la resolución de las cefaleas (cinco días en el grupo de tratamiento versus 13 días en el grupo control; valor de p < 0,0001). Los corticosteroides también se asociaron con números más pequeños de participantes que todavía presentaban cefaleas después de un ciclo de tratamiento de dos semanas (9,1% versus 45,5%; valor de p < 0,0001). El número de pacientes que necesitaron repetir la punción lumbar también fue menor en el grupo de tratamiento (12,7% versus 40%, valor de p = 0,002). Hubo una reducción en la mediana del tiempo de administración de analgésicos en los participantes que recibieron corticosteroides (10,5 versus 25,0; valor de p = 0,038). No se informaron efectos adversos de la prednisolona en el grupo de tratamiento.
Conclusiones de los autores
Los corticosteroides ayudan significativamente a aliviar el dolor de cabeza en pacientes con meningitis eosinofílica que tienen una puntuación de dolor de 4 o más en una escala analógica visual. Sin embargo, hay solamente un ECA que apoya este efecto beneficioso y este ensayo no mencionó claramente la ocultación ni la estratificación de la asignación. Por lo tanto, el estudio incluido se considera un ensayo de calidad moderada. Se necesitan más ECA con diseños adecuados.
PICOs
Resumen en términos sencillos
Corticosteroides para el tratamiento de la meningitis eosinofílica parasitaria
Pregunta de la revisión
¿Los corticosteroides reducen la inflamación de la membrana del cerebro causada por los parásitos?
Antecedentes
La meningitis eosinofílica es una inflamación de la membrana que cubre el cerebro y que se puede clasificar ampliamente en origen infeccioso y no infeccioso. Entre las etiologías infecciosas, Angiostrongylus cantonensis, un gusano del pulmón de la rata, es la causa principal de la meningitis eosinofílica. Esta infección ocurre principalmente en Asia Sudoriental y en toda la cuenca del Pacífico. Sin embargo, este parásito se ha difundido progresivamente más allá de la cuenca del Pacífico y actualmente se encuentra en regiones de Norteamérica debido a la diseminación intercontinental de las ratas de barco infectadas. El dolor de cabeza intenso, que desaparece espontáneamente, es la queja principal. El dolor de cabeza probablemente se debe a una respuesta inmunitaria a los parásitos muertos. Otros signos y síntomas incluyen rigidez y dolor del cuello, trastornos visuales, náuseas, vómitos, parestesia e hiperestesia. Los corticosteroides son fármacos que reducen la inflamación que puede ocurrir en la meningitis eosinofílica debido a las larvas muertas.
Características de los estudios
Se realizó una revisión sistemática y un metanálisis de los ensayos controlados aleatorizados de corticosteroides para el tratamiento de la meningitis eosinofílica. La evidencia está actualizada hasta diciembre de 2014. Esta revisión solamente encontró un ensayo controlado aleatorizado que cumplió los criterios establecidos. Este ensayo incluyó 129 pacientes (63 en el grupo de tratamiento con prednisolona 60 mg/día, dividida en tres dosis durante dos semanas y 66 en el grupo control con placebo). Sin embargo, se perdieron 19 pacientes durante el seguimiento.
Resultados clave
Este ensayo indicó que la mediana del tiempo de resolución del dolor de cabeza fue inferior en el grupo tratado con prednisolona (10,5 días versus 25 días) y el número de pacientes que todavía presentaban dolor de cabeza después de 14 días fue inferior en el grupo de prednisolona en comparación con el control (9,1% versus 45,5%). Hubo diferencias estadísticamente significativas que favorecieron al grupo de tratamiento en otros resultados que incluyeron la frecuencia de administración de acetaminofén (paracetamol) (mediana del número de veces administrado) entre los pacientes que todavía presentaban dolor de cabeza después de 14 días de tratamiento con prednisolona y la media del tiempo hasta la desaparición completa del dolor de cabeza. El número de pacientes que necesitó una repetición de la punción lumbar fue más pequeño en el grupo de tratamiento. No se informaron efectos adversos de la prednisolona en el grupo de tratamiento. Los corticosteroides ayudan significativamente a aliviar el dolor de cabeza en pacientes con meningitis eosinofílica que tienen una puntuación de dolor de 4 o más en una escala analógica visual.
Calidad de la evidencia
Debido a la falta de ocultación de la asignación y el cegamiento (especialmente en un ensayo con resultados subjetivos), y al desgaste (pérdida de participantes), la evidencia se consideró de calidad moderada.
Authors' conclusions
Background
Description of the condition
Eosinophilic meningitis is defined by the presence of greater than or equal to 10 eosinophils per microlitre of cerebrospinal fluid (CSF), or greater than or equal to 10% of the total CSF leukocyte count (Kuberski 1981). The causes of eosinophilic meningitis can be broadly categorised as infectious and non‐infectious. Where the cause is non‐infectious, the aetiologies are leukaemia or lymphoma with central nervous system (CNS) involvement (Hodgkin's), idiopathic hypereosinophilic syndrome (Moore 1985), allergic reactions of the meninges to ventriculoperitoneal shunt (Kennedy 1988; Tung 1991), medications such as non‐steroidal anti‐inflammatory drugs (NSAIDs), ciprofloxacin, trimethoprim‐sulfamethoxazole (Asperilla 1989; Patey 1998; Quinn 1984), and intraventricular vancomycin or gentamicin (Grabb 1992). Intravenous drug injections among drug users have also been shown to cause sterile eosinophilic meningitis and arachnoiditis (Rossetti 2002). When the cause is infectious, the pathogens can be parasites, viruses, bacteria or fungi, with parasites being the most common pathogens. The three predominant agents are Angiostrongylus cantonensis (A. cantonensis), Gnathostoma spinigerum (G. spinigerum) and Baylisascaris procyonis (B. procyonis). Among these, A. cantonensis, the rat lung worm, is the principal aetiologic agent of human eosinophilic meningitis.
In A. cantonensis, humans are infected accidentally by ingesting raw infected molluscs, vegetables contaminated with mollusc slime and carrier hosts such as freshwater prawns, crabs, frogs and planaria (Tsai 2004). This parasite has spread progressively throughout the Pacific basin and is now found in regions of the United States due to intercontinental dissemination of infected ship rats (Campbell 1988; Kliks 1992). The penetration of this host‐parasite system into the tropical and subtropical areas of Africa, the Indian subcontinent, the Caribbean and the temperate Gulf Coast region of the United States is of considerable public health importance.
Eosinophilic meningitis occurs principally in South‐East Asia and throughout the Pacific basin. People in these areas or travellers who eat local uncooked food are at risk of contracting this disease. According to studies in Thailand, there are approximately 1000 new cases every year, most of them among the working‐aged population. The incubation period ranges from six to 31 days following ingestion. Severe headache is the main complaint. Although the headache is self limiting, it is a distressing symptom for the patient. Other signs and symptoms include neck stiffness and pain, visual disturbances, nausea, vomiting, paraesthesia ('pins and needles') and hyperaesthesia (Slom 2003; Tsai 2004). Other signs and symptoms are also self limiting; paraesthesia can persist for weeks but does resolve with time. Paralysis of the facial and extraocular muscles has been reported but generally resolves spontaneously (Kuberski 1979; Podwall 2004).
Description of the intervention
Eosinophilic meningitis is primarily treated supportively with analgesics, sedatives and lumbar punctures to relieve high CSF pressure. These interventions can lead to dramatic clinical improvements. The natural steroid cortisol, a main glucocorticoid, is secreted by the adrenal cortex. Synthetic steroids include prednisolone, methylprednisolone, betamethasone, dexamethasone and triamcinolone hydrocortisone. Steroids can be administered orally, intravenously or by inhalation therapy. Some reports claim a benefit with steroids for eosinophilic meningitis (Koo 1988; Reid 1984), whereas some studies in Thailand have shown no such benefits (Chotmongkol 2002; Punyagupta 1975). Recently, a randomised, placebo‐controlled trial conducted in Thailand demonstrated the effectiveness and safety of steroids for eosinophilic meningitis (Chotmongkol 2004).
Antihelminthics are a class of antiparasitic drugs. Since the main parasites for eosinophilic meningitis are roundworms or nematodes, the antiparasitics used are antinematodes, such as mebendazole, albendazole and thiabendazole. Specific antihelminthic treatment alone is controversial. A study has shown the benefits of combining albendazole and corticosteroids (Chotmongkol 2000). However, the outcome is similar to treatment with corticosteroids alone (Chotmongkol 2009).
How the intervention might work
Steroids act by inducing an anti‐inflammatory response in the body. They are used to treat inflammatory diseases such as arthritis, colitis, dermatitis and bacterial meningitis. Dead larvae cause an inflammatory reaction and exacerbate symptoms, therefore steroids can be used to regulate this natural response. However, the use of steroids can cause adverse effects such as skin lesions, weight gain, Cushing's syndrome, cataracts, osteoporosis, hyperglycaemia, gastrointestinal bleeding, slow healing and psychosis immune suppression. Most of these adverse effects are dose‐ and duration‐dependent.
Why it is important to do this review
The benefit of steroids in the treatment of eosinophilic meningitis remains unclear, although they are sometimes used in clinical practice. This review aims to resolve this question.
Objectives
To assess the efficacy and safety of corticosteroids for the treatment of eosinophilic meningitis.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs) of corticosteroids for eosinophilic meningitis.
Types of participants
Participants aged 15 years or older, of any sex, with eosinophilic meningitis. The diagnosis of eosinophilic meningitis is usually made from a medical history, dietary history, clinical signs and symptoms, and cerebrospinal fluid (CSF) analysis. Imaging of the brain is sometimes used to exclude other potential causes. We excluded participants if they were already taking steroids at the time of the study.
Types of interventions
Steroids (any type and dosage) versus placebo.
Types of outcome measures
Primary outcomes
-
Resolution rates (complete disappearance of headache within two to four weeks after completion of the treatment).
Secondary outcomes
-
Percentage of analgesics used to relieve symptoms.
-
Percentage of therapeutic lumbar punctures used to relieve headaches.
-
Time to complete resolution of headache after treatment.
-
Adverse effects of medication.
Search methods for identification of studies
Electronic searches
For this 2014 update of we searched the Cochrane Central Register of Controlled Trials (CENTRAL 2014, Issue 11) (accessed 10 December 2014), which contains the Cochrane Acute Respiratory Infections (ARI) Group's Specialised Register, MEDLINE (May 2012 to November week 3, 2014), EMBASE (June 2012 to December 2014), Scopus (2012 to December 2014), Web of Science (2012 to December 2014), LILACS (2012 to December 2014) and CINAHL (May 2012 to December 2014).
Previously we searched CENTRAL (2012, Issue 6), MEDLINE (1950 to July Week 4, 2012), EMBASE (1974 to July 2012), Scopus (1960 to July 2012), Web of Science (1955 to July 2012), LILACS (1982 to July 2012) and CINAHL (1981 to July 2012).
We used the search strategy in Appendix 1 to search MEDLINE and CENTRAL. We combined the MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity‐maximising version (2008 revision); Ovid format (Lefebvre 2011). We adapted the search strategy to search EMBASE (Appendix 2), Scopus (Appendix 3), Web of Science (Appendix 4), LILACS (Appendix 5) and CINAHL (Appendix 6). We imposed no language or publication restrictions.
Searching other resources
We searched the trials registries World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) and ClinicalTrials.gov (latest search 29 July 2014) for completed and ongoing trials. We identified other relevant trials by searching the reference lists of included trials and handsearching conference abstracts and non‐indexed articles.
Data collection and analysis
Selection of studies
Two review authors (SiT, SaT) independently selected studies from the search results. We obtained the full text of the article in case we were in doubt about the trial's eligibility. We resolved any disagreement between the review authors by discussion.
Data extraction and management
Two review authors (SiT, SaT) independently extracted data from the studies using a predefined protocol. We resolved any disagreement between the review authors by discussion. Two review authors (SiT, SaT) independently collected data from the potential studies. One review author (CN) entered the information from the data extraction forms and analysed the quantitative data using Review Manager software (RevMan 2014).
Assessment of risk of bias in included studies
We assessed risk of bias in the included studies across six domains (sequence generation, allocation concealment, blinding of participants, blinding of personnel, blinding of outcome assessors, incomplete outcome data, selective outcome reporting, and other potential sources of bias). Two review authors (SiT, SaT) judged each domain as low risk of bias, high risk of bias, or unclear risk of bias, based on the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). In case of disagreement, the third review author (CN) acted as arbitrator.
Measures of treatment effect
We measured the treatment effect of the studies using the risk ratio (RR) with 95% confidence intervals (CIs) for dichotomous outcomes (percentage of patients who still had a headache after two weeks completion of treatment, analgesics used to relieve symptoms, therapeutic lumbar puncture to relieve headache and adverse effects of medications).
We used the mean difference (MD) or standardised mean difference (SMD) with 95% CIs for continuous outcomes (time to complete disappearance of headache after treatment).
Unit of analysis issues
The unit of analysis was the individual participant.
Dealing with missing data
We contacted the authors of the included studies by email to request missing data. In case of missing participants due to drop‐out, we used an intention‐to‐treat (ITT) analysis.
Assessment of heterogeneity
We did not investigate the heterogeneity of treatment effect between studies as there was only one study that met our inclusion criteria.
Assessment of reporting biases
We did not assess reporting bias.
Data synthesis
As there was only one included study, we were unable to pool the treatment effects.
Subgroup analysis and investigation of heterogeneity
We did not conduct a subgroup analysis to investigate the source of heterogeneity.
Sensitivity analysis
We did not carry out sensitivity analysis.
Results
Description of studies
See: Characteristics of included studies and Characteristics of excluded studies tables.
Results of the search
We retrieved 21 records from this 2014 update search but we did not identify any new trials for inclusion or exclusion. Previously we retrieved 90 records from our search of the electronic resources. We found one eligible trial (Chotmongkol 2000). We also identified one trial that was close to meeting our inclusion criteria but the study compared the treatment effects of prednisolone (corticosteroid) plus albendazole (antihelminthic) with prednisolone alone, not with placebo (Chotmongkol 2009).
Included studies
Only one study was eligible for inclusion (Chotmongkol 2000). In this study, the participants were patients aged 15 years or older with a diagnosis of eosinophilic meningitis based on cerebrospinal fluid (CSF) findings of ≥ 10% eosinophils. One hundred and twenty‐nine participants (63 in the treatment group and 66 in the control group) were enrolled. However, 19 participants (eight in the treatment group and 11 in the control group) were lost to follow‐up and data were incomplete.
The participants, who were stratified according to the severity of headache and CSF opening pressure, were randomised to receive either treatment or placebo. The intervention consisted of prednisolone 60 mg/day divided into three doses for two weeks. During the treatment, patients were given two tablets of acetaminophen (500 mg tablet) every four to six hours to relieve headache; repeat therapeutic lumbar punctures were conducted if the headache was not resolved using acetaminophen.
The trial authors recorded the frequency of acetaminophen use and the percentage of patients who needed lumbar punctures. Both groups also took alum milk orally, divided into three doses after meals. Participants were evaluated every day for two weeks, then every two weeks until they completely recovered. The severity of headaches was assessed using a visual analogue scale. The short‐term adverse effects, such as gastrointestinal bleeding and hyperglycaemia, were assessed. There were no disagreements on the inclusion or exclusion of studies between the review authors extracting study data. We did not need to contact any trial authors to provide additional information for this version of the review.
Excluded studies
We excluded five studies that did not fit our inclusion criteria. Four studies were not RCTs (Chotmongkol 2004; Chotmongkol 2006; Chotmongkol 2009; Sawanyawisuth 2004); and one study compared prednisolone plus albendazole with prednisolone alone, not with placebo (Chotmongkol 2009).
Risk of bias in included studies
The overall risk of bias is presented graphically in Figure 1 and summarised in Figure 2.
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
Allocation
In our included study, the sequence generation used a block‐of‐four randomisation and was therefore adequate (Chotmongkol 2000). However, allocation concealment was not mentioned in this study, which can lead to selection bias (Schulz 2002).
Blinding
Chotmongkol 2000 used a placebo as a control and so the participants and investigators, including outcome assessors, were blinded.
Incomplete outcome data
Nineteen participants were lost to follow‐up (eight in the treatment group and 11 in the control group); the reasons were not adequately described in the study. In the worst‐case scenario, if the missing participants in the treatment group still had headaches after 14 days, the number of participants who still had headaches would be 13/63 (20%); and for the control group it would be 25/66 (37%).
Selective reporting
There was no selective reporting of outcomes in this study. The trial authors described important outcomes mentioned in the methods of the study.
Other potential sources of bias
The baseline characteristics were not significantly different between the two groups in the study. However, the assessment of drug compliance was not mentioned in the results.
Effects of interventions
Primary outcomes
1. Resolution rates (complete disappearance of headache within two to four weeks after completion of the treatment)
There was a statistically significant difference in the outcomes. The number of patients who still had headaches after 14 days of prednisolone treatment was significantly higher in the control group (25 out of 55 (45.5%) versus 5 out of 55 (9.1%), risk ratio (RR) 0.2, P value = 0.00004). The mean time until complete disappearance of headache was significantly shorter in the treatment group (5 days versus 13 days, P value < 0.001).
Secondary outcomes
1. Percentage of analgesics used to relieve symptoms
The frequency of acetaminophen use (median of number of times) was lower in the treatment group: 10.5 versus 25, P value = 0.038.
2. Percentage of therapeutic lumbar punctures used to relieve headaches
The number of patients who needed repeat lumbar puncture was smaller in the treatment group: 7 (12.7%) versus 22 (40%), RR 0.32, P value = 0.002.
3. Time to complete resolution of headache after treatment
The (median) time to complete resolution of headache after treatment was shorter in the treatment group: 5 (1 to 60) days versus 13 (1 to 56) days, P value < 0.0001.
4. Adverse effects of medication
There were no reported adverse effects from prednisolone in the treatment group.
Discussion
Summary of main results
The use of corticosteroids (prednisolone 60 mg/day) showed a beneficial effect in parasitic eosinophilic meningitis. The single study we included showed that a two‐week course of prednisolone 60 mg/day in three divided doses shortened the length of headaches caused by eosinophilic meningitis, reduced the frequency of analgesic use and reduced the number of patients who needed therapeutic lumbar puncture (Chotmongkol 2000). There were no adverse effects from steroids reported in either group.
Overall completeness and applicability of evidence
According to our review, based on one included study, a two‐week course of prednisolone 60 mg/day in three divided doses was beneficial in adult patients aged ≥ 15 years who had eosinophilic meningitis, diagnosed by ≥ 10% eosinophils in the cerebrospinal fluid (CSF) (Chotmongkol 2000). All patients in the included study had a history of snail ingestion prior to the onset of eosinophilic meningitis. This study was conducted in Thailand, which is a low‐income country; therefore, the role of corticosteroid use for eosinophilic meningitis in high‐income countries is not clear. However, most eosinophilic meningitis occurs in low‐income countries, since one of the risk factors is food hygiene.
For prednisolone use in relation to the severity of headache, we found that all patients in this study had at least a moderate degree of headache. Thus we assumed that prednisolone would be beneficial in treating moderate to severe headaches (4 to 7, moderate pain; 8 to 10, severe pain), but the role of corticosteroids in treating mild headaches needs to be addressed in future studies.
There were no adverse events reported in either group. No gastrointestinal bleeding or hyperglycaemia were reported in either group. The duration was not specified in the study but we could assume that there were no such adverse events for at least two weeks during the study period. Another study has compared the benefit of prednisolone alone versus prednisolone plus albendazole in the treatment of eosinophilic meningitis, but as this trial did not compare prednisolone plus albendazole versus placebo, we did not include this study in our review (Chotmongkol 2009). However, there were some points worth mentioning from this study. It did not show any advantages of prednisolone plus albendazole over prednisolone alone, leading to a conclusion that prednisolone plus albendazole was not superior to prednisolone alone in the treatment of eosinophilic meningitis.
Quality of the evidence
Given the lack of allocation concealment and blinding (especially in a trial with subjective outcomes), the attrition and the significant differences in outcomes, we agreed to grade our included study as a moderate quality trial (Chotmongkol 2000).
Potential biases in the review process
We did not identify any potential biases in the review process.
Agreements and disagreements with other studies or reviews
There have been no previous systematic reviews regarding the role of corticosteroids in eosinophilic meningitis. One prospective cohort study showed a benefit of a one‐week course of 60 mg/day of prednisolone to relieve headaches caused by eosinophilic meningitis (Sawanyawisuth 2004). Another study conducted in Taiwan, which was a retrospective cohort study, showed that treatment with mebendazole (antihelminthic) plus dexamethasone (corticosteroid) shortens the duration of illness compared to those treated with analgesics alone, but the authors did not mention the dose of dexamethasone (Tsai 2001). In contrast, another study did not show any advantages of steroids in the treatment of eosinophilic meningitis (Punyagupta 1975). There were no adverse effects from corticosteroid use in the aforementioned studies.
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
