2.1. Study Design

A descriptive, survey-based cross-sectional and multicenter study was conducted to determine the characteristics of access to diagnosis and treatment of AE in Latin American and Caribbean countries. The study was carried out from October to November 2023.

2.2. Population and Sample

This study was conducted through the REAL LABIC project, which includes all medical professionals involved in the care of patients with AE in Latin America and the Caribbean. Its main objective was to understand the different aspects of AE in our region, especially its epidemiology, clinical characterization, and management [12]. Once the participating centers were selected, a specialty doctor in charge of providing information on the strengths and weaknesses in AE diagnosis, treatment, and follow-up at their hospital acted as a representative. The physician representing the center was selected through a prior interview with the center’s participants as part of the REAL LABIC project. Following a training process on the key variables of the study, participants chose their representative, considering that the selected individual had to be responsible for the care of patients with AE. The sample size was not calculated for this study since we used the entire sampling frame available for a census study.

2.3. Instruments and Variables

A questionnaire was created using Google Forms and distributed to each selected participant, where relevant information was collected such as characteristics of the specialty (age, years of clinical practice, specialty, age group management in patients with AE, number of patients with AE treated, and type of hospital); access to diagnosis (type of electroencephalogram (EEG) monitoring performed, neuroimaging available at their center, processing of autoantibodies at their hospital, processing time for antibodies, assuming costs of antibody testing for AE, and expected time to evaluate response to treatment in AE); and the treatment and management of encephalitis (initiation of AE treatment, maintenance therapy used, use of prednisone after initial treatment, use of protocol for management, training in management, and reasons for starting management).

2.4. Statistical Aspects

Data were stored in a Microsoft Excel spreadsheet, subsequently imported, and analyzed using the statistical program STATA v16.0 (Stata Corp, Texas, United States). In the descriptive analysis, categorical variables are presented as frequencies and percentages, and measures of central tendency and dispersion are shown for quantitative data. Based on the number of participating countries, a distribution map was created (Figure 1).

2.5. Ethical Aspects

This research study had ethical approval from the project “Disparities in access to diagnosis and treatment of AE in Latin American hospitals” prior to its execution, with university research registration (315-CEI-EPM-UCV-2023). No personal patient data were used to carry out the study, and no sensitive information was involved in patient management; however, the health personnel provided us with routine information on the diagnosis and treatment of AE at their hospital centers.

3.1. General Characteristics of Participants

A total of 108 physicians participated, representing 19 countries in Latin America and the Caribbean (Figure 1). Participants’ median age and years of medical practice were 40 and 13 years, respectively. Regarding specialties, individuals who responded the most to the survey were general intensivists (31.5%), neurologists (28.7%), and neurointensivists (17.6%). A little over half of the participants worked in public centers (55.6%) and three-fourths of all centers were university hospitals. The majority of respondents (65.1%) identified neurology as the specialty responsible for managing patients diagnosed with AE. Most respondents had experience managing cases of AE in adults (76.9%). Regarding the number of patients evaluated and managed by participating doctors, 54.2% had at least experience managing between 1 and 5 patients, and only 6.5% of those who responded to the survey had managed more than 20 patients (Table 1).

3.2. Access to AE Diagnosis in Latin America and the Caribbean

Among those surveyed, 84.3% had access to at least one modality of EEG (continuous or quantitative EEG), and 15.7% did not have access to EEG methods. All hospitals evaluated had at least one neuroimaging modality, 82.5% had access to contrast-enhanced brain magnetic resonance imaging (MRI), and 17.6% had access to contrast-enhanced tomography. Approximately 15% could perform positron emission tomography. All evaluated hospitals had access to performing lumbar punctures and the corresponding tests (cell count, protein levels, oligoclonal bands, and microbiological testing). Out of the hospitals included, 72.7% did not process antibodies in serum and/or cerebrospinal fluid; however, in 36.1% of cases, antibodies were processed nationally, and 23.1% of the tests were processed in laboratories outside the country where the samples were collected. The processing time was between 1 and 20 days in 78.7% of patients and > 20 days in 21.3%. Only 3.7% of tests took 30 days to report the results. The costs for conducting the tests were paid by the patient in 44.4% of cases and in 31.5% of cases by the health system. The remaining 24.1% were paid by private insurers, the laboratory, the hospital, or co-financed (Table 2).

3.3. Access to Treatment and Management of AE in Latin America and the Caribbean

Regarding respondents’ preference for the initial treatment, 52.3% preferred using methylprednisolone as the only treatment and 27.1% used dual therapy as initial therapy in cases of AE. If there was a lack of therapeutic response to the initial treatment, 43.9% added another first-line agent and 36.4% escalated to second-line therapy with rituximab. Regarding the occasional use of a third-line therapy, 16.6% used at least one agent; the drugs of choice being tocilizumab (15.7%) and bortezomib were used in only one case.

In terms of maintenance therapy, one-fifth of the participants did not specify the one used, and the rest, when indicated, preferably used prednisone (49.5%), rituximab (17.8%), and monthly intravenous immunoglobulin (6.5%). Regarding prednisone bridging therapy after first-line treatment, 53.3% of participants used it if the first-line treatment included methylprednisolone, and 19.6% indicated it even if it had not. Furthermore, 85.2% of participants started initial treatment within the first 10 days of patients’ admission. Regarding the delay in starting treatment, most participants mentioned initial suspicion of another disease (44.4%) or nonavailability of antibodies at their hospital (27.8%). Additionally, regarding the approximate number of patients with AE estimated to have continued outpatient follow-up by a specialist, most participants (39.8%) estimated that less than 10% of patients had follow-up.

Moreover, regarding whether their hospital has a written protocol for managing patients with AE, 89.8% did not have one, and 70.8% of participants had not received any institutional refresher course on managing this condition (Table 3). Finally, as for the drugs used to control epileptic seizures or status epilepticus, those with the greatest availability were propofol (97.1%), intravenous phenytoin or oral valproate (96.3%), and midazolam (96.2%) (Table 4).

4.1. Diagnosis and Access to Auxiliary Studies

The diagnosis of AE is challenging for clinicians because of the different clinical conditions that can present with similar symptoms (especially infectious encephalitis in tropical climates) and the availability of adequate diagnostic methods [13]. Therefore, a diagnosis based on the combination of a detailed clinical history, exhaustive physical and neurological examinations, and appropriate use of diagnostic aids is very important. The criteria for the diagnosis of AE published in 2016 [4] establish the framework for an accurate diagnosis and include performing brain MRI, cerebrospinal fluid studies, EEG, measurement of neural autoantibodies, and, in selected cases, the search for primary tumors. In our investigation, access to brain MRI was 82.5%. This suggests that one in five hospitals evaluated does not have access to this imaging modality. In all cases, however, there is access to simple and contrast-enhanced tomography, the usefulness of which is very limited in this context but helpful in some differential diagnoses. Positron emission tomography is available in 15% of health care centers, and its usefulness in AE is increasingly being described. It can show earlier changes than MRI, have greater sensitivity in early stages, and can help search for related tumors [14]. Therefore, there is universal access to neuroimaging at the centers studied, but there are limitations in performing brain MRI and positron emission tomography according to the currently available recommendations [15, 16].

The timely measurement of antineural antibodies in suspected patients from the environment studied was poor, and significant difficulties were experienced. Only 27.8% of the centers evaluated were able to perform the test at their institution. The methods for conducting this measurement consisted of sending the sample to laboratories within the same country (68.5%) or abroad (23.1%). One in five hospitals reported more than 20 days between test collection and results, and a considerable minority (27%) of respondents delayed treatment until antibody results were available, suggesting that treatment may be delayed in some cases due to delays in antibody testing results. Previous investigations in our region show low availability of antibody testing, delays in treatment, higher mortality, and associated sequelae [17, 18]. The difficulties in our study have been widely described in other studies and reports, specifically in Brazil, through the Brazilian Autoimmune Encephalitis Network [19].

One of the most significant barriers found in our study was poor accessibility to funding for antineural autoantibody testing. It is concerning that 44.4% of patients had to bear the test cost. Previous reports estimate costs between 300 and 900 US dollars, constituting one and three times the minimum monthly income in the regions evaluated [20, 21]. Public health systems assume this cost in 31.5% of the centers evaluated; however, in cases like Colombia, only the dosage of anti-N-methyl-D-aspartate (NMDA) antibodies is covered [22]. The availability of antibody tests in other low-income regions of the world has been paid for by private laboratories or the patient in most cases [20]. In our investigation, private laboratories and hospital financing represented only 3.8% of the sources. Out-of-pocket spending on health represents inequity and a perpetuator of the cycle of poverty and disease, especially in low- and middle-income areas [23].

EEG studies are suggested in international guidelines for studying patients with AE [15, 16]. Their usefulness lies in the fact that they can identify focal abnormalities with normal MRI results, help diagnose specific types of AE, and are, for example, part of the diagnostic criteria for limbic encephalitis. The most common findings are nonspecific (slowing and focal discharges). Extreme delta-brush is seen in 30% of anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis and other specific types. In our investigation, 84.3% of participants had access to at least one modality of EEG (continuous or quantitative EEG). Thus, we consider that, although most centers having the power to perform EEG studies is a strength, the fact that 15% of hospitals did not have the possibility of performing any type of EEG is critical and represents an additional challenge to managing the complexity of this disease.

4.2. Treatment and Clinical Follow-Up Protocols

It was evident that most participants in our study reported not having institutional protocols (89.8%) and had not received any academic training from their institutions (70.8%). This is something important that must be reinforced in hospitals in low- and middle-income countries, where the availability of tests and treatment is limited since an incorrect diagnosis of AE can increase morbidity due to unnecessary immunotherapy and delay the proper management of these patients [24]. Furthermore, it was reported that most participants started treatment < 10 days after the patient’s admission to start some immunosuppressive treatment and < 7 days to indicate treatment escalation in case of lack of therapeutic response. This finding denotes the importance given to AE when it is suspected since timely diagnosis and treatment are associated with a better functional prognosis. However, international good practice recommendations indicate that the exact time to evaluate therapeutic response is poorly defined [25].

Regarding access to treatment, most participants could access first-line immunosuppressive treatment, and there was growing access to second-line therapies. However, it is important to mention that in some countries, certain drugs, such as immunoglobulin or rituximab, are not regulated or available as a therapeutic option [8], although there is some evidence in patients with some types of AE. Also, immunoglobulin is one of the most used first-line agents, and rituximab is the second-line agent most used in other countries with an adequate level of efficacy and safety [25–29].

Methylprednisolone was reported to be the preferred agent for most participants. However, a trend was also seen toward the use of first-line combination therapy as a first intention or as step-up therapy in the event of failure of the use of a first-line first agent. This is recommended in cases of AE considered severe (NMDAR and severe dysautonomia) [25]. Furthermore, 79.6% reported not requiring third-line treatment, which could reflect the effectiveness of first- and second-line agents for managing patients with AE.

There was heterogeneity in the indication for maintenance treatment, with corticosteroid therapy, rituximab, and immunoglobulin being the most used therapeutic agents. The preferred use of prednisone as bridging therapy among the survey participants was remarkable, and it has been previously reported as bridging immunotherapy to discharge [30]. Likewise, there was also heterogeneity in the availability of the use of antiseizure medications, but most centers had the availability of at least one intravenous medication and one oral medication for seizure control in these patients. This is important because epileptic seizures are one of the main clinical manifestations of AE and are the reason for admission to critical care units [31].

Regarding clinical follow-up, only 10% of participants reported outpatient follow-up of treated patients diagnosed with AE. Follow-up of these patients after the initial episode is as important as the treatment in the acute phase of the disease since it is widely known that a percentage of patients can relapse. The appearance of neoplasms in the future, or physical and neuropsychological sequelae, has also been reported [30]. In our setting, cognitive sequelae have been described in > 90% of patients beyond 1 year of follow-up [32, 33].

4.3. Continuing Medical Education

Despite an average experience of 13 years among the personnel evaluated, the majority (93.5%) had treated < 20 patients with AE. This suggests that, on average, three patients are seen over 2 years. This is consistent with the incidence of the disease and the absence of specialized centers for these patients, which generate geographic dispersion for treatment. University education and continuing medical training are of major importance in this scenario of low exposure to real patients. However, despite having 73% of university hospitals among those evaluated, 70% of the personnel had not received training courses in AE from their institutions in the last year. Additionally, strikingly, < 1% of those evaluated had specialized training in neuroimmunology, and therefore, the presence of subspecialized medicine does not make up for the training deficiencies of the basic specialties. Therefore, a greater presence of specialized medicine in areas of neuroimmunology is required, as well as greater training in basic interdisciplinary specialties in charge of patients with AE in our environment. In the future, as Moeller et al. note, there will be an increase in emphasis on educational outcomes and the public impact of neurology training programs [34]. In summary, although AE requires specialized hospital infrastructure, training human talent is vital for environments with limited resources and growing concerns about cost-effectiveness and cost-usefulness for the design of public policies.

4.4. Future Perspectives

A recent bibliometric analysis showed how most publications on AE were conducted in high-income Western countries [35]. This creates a bias for the design of clinical protocols, specific recommendations, and public policy planning for the regions and their particularities not represented in these studies. Bhatia et al. provide recommendations to reduce the impact of these gaps, namely, improvement in training for the identification of classic syndromes as well as the red flags such as hyponatremia and cardiac arrhythmias, administration of empirical therapy in cases of high suspicion even without confirmation through serological or cerebrospinal fluid, and enhanced development of multidisciplinary protocols and policies given the reversibility of these entities and the performance of regional clinical studies [20].

4.5. Authors’ Recommendations to Improve Access to Patients With AE in Resource-Limited Settings

4.6. Strengths and Limitations

Public and university institutions in > 15 countries in Latin America and the Caribbean participated, which provides a broad overview of the situation in the region. The study included many specialties involved in caring for patients with AE, providing a global and interdisciplinary perspective. Participants consisted of medical personnel directly involved in the care of these patients, with a high average of years of clinical practice and who worked in reference hospital centers in the different countries of the region studied.

Our study has several limitations. Regarding the respondents, this study reported that fewer than one-third were neurologists; however, neurology was identified as the specialty responsible for managing AE in nearly two-thirds of the surveyed hospitals. This may have influenced some of the results related to the number of AE patients treated by the respondents. This represents a limitation of the study and suggests that the inclusion of more neurologists might have provided a more accurate picture of how AE cases are managed. Since it was carried out through a survey, the results are exposed to memory and selection biases because the survey cannot reach all the actors involved. Given the absence of institutional protocols and scientific societies, the specialists in charge are heterogeneous and may have “school protocols” that affect the uniformity of the data. No information was obtained on the direct cost of antibody testing, nor was it possible to establish the techniques by which autoantibodies are measured in local laboratories (for example, cell-based assay, immunoblotting, radioimmunoassay, and enzyme-linked immunosorbent assay), and the importance of the technique used related to diagnostic performance is well known.