Clinical Characteristics and Predictors of Short-Term Outcome in Mexican Adult Patients with Guillain-Barré Syndrome
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.310063
Source of Support: None, Conflict of Interest: None
Keywords: Acute inflammatory demyelinating polyneuropathy, acute motor axonal neuropathy, acute neuromuscular failure, Guillain-Barré, peripheral neuropathy
Guillain-Barré syndrome (GBS) is an acute inflammatory polyradiculoneuropathy characterized by flaccid areflexic paralysis with an autoimmune basis. From a global standpoint, GBS occurs with an incidence of one or two cases per 100,000 subjects per year., Although the consensus is to consider GBS as a single disorder with clinical variability typically marked by inflammatory demyelination, past evidence leads us to classify using demyelinating or axonal categories according to nerve conduction studies. Pathological diagnosis is infrequently used in practice.,
In Mexico, isolated reports with a small number of GBS patients derived from several hospital reference centers have almost consistently found a predominance of the axonal pattern with a mortality rate of 10.5% with scarce information regarding acute treatment,, In this study, we include the clinical presentation, electrophysiological study findings, acute treatment choice, and short-term prognosis.
We considered unbiased consecutive patients older than 18 years of age from January 2002 to February 2011, with a diagnosis of acute neuromuscular failure compatible with GBS according to Asbury criteria. These patients were admitted to the neurology, internal medicine, or intensive care unit (ICU) departments at Hospital Civil de Guadalajara Fray Antonio Alcalde.
Patients with the Miller-Fisher variant and recurrent GBS (patients that had two or more episodes of GBS, with either a minimum interval >4 months between episodes or patients that did not recover completely at >2 months when there was a complete or near-complete recovery after the previous episode) were excluded. Those with a previous disability for other medical conditions were also excluded.
Clinical data regarding sociodemographic profile, previous infections, other comorbidities, course, clinical presentation, days from symptom onset to hospital admission, neurological examination, treatment, hospital setting care, and discharging hospital conditions were also registered. Mandatory nerve conduction studies and optional lumbar punctures were obtained within 15 days from hospital admission. Cerebrospinal fluid (CSF) compatible with GBS diagnosis included an increase in proteins (preferably after the first week of symptom onset) in the absence of pleocytosis (>50 cells/ml).
The degree of disability was evaluated at hospital admission based on the Hughes functional scale as follows; 0: healthy, 1: minor signs or symptoms of neuropathy but capable of manual work/running, 2: able to walk without the support of a walking stick but incapable of manual work, 3: able to walk with a walking stick, mobility aid, or support, 4: confined to bed or chair bound, 5: requiring assisted ventilation; and 6: dead.
We considered a diagnosis of mild GBS in those patients with a minimum disability according to Hughes scale 0, 1, and 2, and as a severe GBS, those with a Hughes of 3, 4, and 5 at hospital admission. Nerve conduction velocities were performed after the first seven days of clinical onset, using standard procedures. The electrophysiological variants, according to Hadden's criteria, were considered as follows: acute inflammatory demyelinating polyneuropathy (AIDP), axonal forms as acute motor axonal neuropathy (AMAN) and acute motor and sensory axonal neuropathy (AMSAN). We also included the equivocal form when it was not possible to classify the findings in some of the groups mentioned above or had a mixed axonal and demyelinating pattern. Functional outcome at hospital discharge, at 3 and 6 months, was also evaluated using the Hughes scale. Good outcome at discharge was considered using a score of 0–2, and poor outcome using a 3–6 score in the Hughes scale. At 3 and 6 months, a good outcome was considered using a score of 0–1, and poor outcome as 2–6 using the Hughes scale. Pain complaints were defined as any GBS symptoms presentation of radicular, neuropathic, arthralgic, or myalgic aching complaints. Early gait complaints were defined as any feeling of fatigue, weakness, numbness, or postural instability with final gait difficulty as onset presentation. Ethical Committee to our hospital approved the study in January 2001. Written consent was obtained in all cases.
Demographic data are presented using measures of central tendency. Pearson's X2 test was used for comparing nominal qualitative variables between two groups, or to evaluate the homogeneity in the distribution of these variables in three or more groups. Classification variables analysis was performed with the use of Chi-square test. Univariate analysis was used to identify factors associated with good prognosis at hospital discharge, which were further analyzed by Wald Test of Logistic regression analysis for predictors independently related to poor prognosis and good prognosis. Survival analyses were calculated during hospital stay; these curves were generated using the Kaplan-Meier estimator and were compared by the log-rank test. All P values less than 0.05 were considered statistically significant. Statistical analysis was carried out using SPSS v. 22 (Chicago, Ill). Local Ethical Committee review board approved the protocol, and informed consent was obtained from each patient, relative, or primary caregiver.
A total of 148 patients with acute neuromuscular failure were considered. Thirty-three were excluded for the following reasons: 11 lacked electrophysiological studies, seven had recent exposure to organophosphates, three corresponded to acute spinal myelopathy, two had a diagnosis of hypokalemic periodic paralysis, 1 due to acute porphyria, and another associated with hepatitis C. Five patients with the Miller-Fisher variant and three patients with recurrent GBS were also excluded.
Of the 115 patients with final GBS diagnosis, 78 were men (68%) and 37 women (32%); the mean age was 44 years (range 18–84 years). Eight (7%) patients were older than 70 years. In total, 63% of patients had a preceding infectious event, either diarrhea or an upper respiratory tract infection (URI). Nineteen (17%) patients came from rural areas and 96 (83%) from urban areas. The main clinical presentation was weakness in 110 (96%) patients, sensory impairment 59 (51%), any pain complaints 42 (37%), early gait complaints 30 (26%), and dyspnea in 31 (27%).
The mean onset from the onset of symptoms to hospital admission was 7 days (range 1–17 days). Nine patients reached the nadir, hospitalized at 14 days from symptom onset. Patients older than 71 years and with a history of URI were significantly more likely to be hospitalized in the first week after symptom onset (P = <0.05).
Classic ascendant GBS form was observed in 75 (65%) patients; however, the descendent pattern was also prevalent and reported in 40 (35%) patients: 30 as descendent cervicobrachial, 6 descendent cranio-cervical (facio-cervicobrachial), and 4 as nonspecific descendent.
At hospital admission, 15 (13%) patients were classified as mild GBS, and severe GBS was found in 100 (87%) patients. Due to the severity of the illness, 23 patients (20%) required ventilatory support and ICU admission. Based on the electrophysiological classification, the frequency of GBS forms was as follows: AMAN in 36 (31%), AIDP in 33 (29%) patients, AMSAN in 21 (18%), and in 25 (22%) results were equivocal. Lumbar puncture was performed in 68 patients (59%) with the CSF finding of albuminocytological dissociation in 55 (81%) cases [Table 1].
From a total of 115 patients, 73 (64%) were treated with induction therapy (93% of patients with Hughes scale 0–2 were not treated). Fifty of the 73 treated patients (68%) received plasmapheresis, 13 (18%) IVIG, and 10 (14%) both (plasmapheresis as initial management). The initial therapy was chosen according to immediate availability, considering IVIG as first-choice therapy. The average hospital stay was 15 days (range from 2 to 72 days). During this time, 14 (12%) patients died within the hospital due to systemic complications, mainly pneumonia or sepsis. In a complementary univariate analysis, main variables significantly associated with death compared to mild or severe Hughes at hospital discharge were age ≥71 years old and dysphagia or dyspnea as a clinical onset manifestation (P < 0.05) [Table 2] or [Supplemental Table].
A history of URIwas significantly present in patients with AIDP (P = 0.037); meanwhile, pain complaints were significantly present in the equivocal GBS group (P = 0.03). At hospital discharge, early gait complains and mild GBS were significant for a good prognosis (all with P < 0.05); dysphagia, dysarthria, dyspnea, and severe GBS at admission were significant for poor prognosis (all with P values <0.05) [Table 3].
We followed 50 patients at 3 months and 33 of these patients at 6 months; a mild GBS at hospital admission (Hughes scale 0–2) and to be discharged with a mild GBS (also with Hughes score 0–2) were significantly associated with a good outcome, while admission to the hospital with a 3–5 Hughes score and to be discharged with a Hughes score >2 was significantly associated with a worse outcome (both with P values = <0.05). At a 6-month follow-up, AIDP pattern was associated with a tendency for a good outcome, and the equivocal form was significantly associated with a worse prognosis (P < 0.05) [Table 4]. In a multivariate binary logistic regression analysis, early gait complaints (OR: 6.22, IC 95%: 1.4–27.6) and hospital admission with a mild Hughes score (OR: 61.25, IC 95%: 11.78–318.5) was predictive for a good outcome at hospital discharge (P < 0.05) [Table 4]. In an in-hospital survival analysis, according to the Kaplan Meier estimate, patients aged <70 years showed better survival than older ones (P < 0.001) [Figure 1]. In patients who reached 3 and 6 months follow-up, a mild Hughes score (0–2) at hospital admission and hospital discharge also with a mild Hughes score (0–2) was associated with a good outcome (Hughes 0–1).
Our findings are similar to those reported in international literature concerning demographic characteristics of GBS, such as mean age and a male predominance. In our site, upper respiratory infection history was significantly associated with the AIDP pattern, as in a recent retrospective analysis. Pain complaints were less common among AMAN patients and more frequent in patients with an equivocal pattern. This finding is according to a Chinese report, where radicular pain was reported in 19% of the AMAN subtype GBS, independently to controversial concerns about not finding the expected symptoms in the axonal-motor form of GBS.,
We confirmed a higher prevalence of the axonal forms of GBS-AMAN/AMSAN, similar to previous reports from our country; a finding that is different from reports in other countries, where the most common variant is AIDP.,,, Interestingly, in Mexico, GBS was first described in 1969 as a neuronopathy subtype different from the classic form of poliomyelitis and also different from the classical AIDP-GBS that affected motor nerves. Biopsies later revealed that the insult was located precisely in the axon. Twenty years later, Mckhann et al. characterized a group of 36 patients from North China with acute motor failure with evidence of axonal motor damage calling it “Chinese paralytic syndrome.” Two years later, biopsy nerves showed that 60% presented axonal degeneration with lymphocytic infiltration, calling the subtype AMAN. Recent studies have shown a higher risk of the AIDP variant associated with specific HLA-II (human leukocyte antigen type II) variants in population-based studies, similar to the data reported in our country where the presence of the allele HLA-DR3 is associated as a strong risk factor. However, no HLA-II associations with the AMAN variant have been found, so further research is required to explain the higher prevalence of this subtype in our population.,
We also notoriously found a higher frequency of descendent pattern of weakness. In other Mexican GBS samples, 17% of patients also presented this descendent pattern.
Clinical variables associated with a benign course in our site was an age <40 years (P = 0.041). Clinical variables associated with mechanical ventilation use after the first week of hospital admission were the time between weakness onset and hospital arrival, the Medical Research Council sum score, and the presence of facial or bulbar weakness.,
Interestingly, neither hospital setting care (general ward or ICU stay) nor any of the conventional induction therapies approved in GBS treatment had an impact at the hospital discharge or short-term follow-up outcome. According to a recent study of 186 GBS patients, the main conclusion was that the long-term prognosis was not influenced by the choice of treatment. In this way, previous reports argue about the use of these treatments because most of these studies were derived from Europe and North America, and, therefore, most of these patients probably had the AIDP form.
In our study, 10 of 115 patients received both treatments due to the initial inadequate response. A previous report from 2010 concluded that the use of plasmapheresis after IVIG did not improve the short-term outcome of patients with GBS and also showed that it increases cost and duration of hospitalization. We also confirmed that the use of both treatments did not show an impact on the patient's clinical course or short-term outcome.
A good outcome at hospital discharge was observed in 16% of our patients, a similar frequency to other reports. In a recent study, it was found that male sex, age >50 years, >14 days onset symptoms to hospital arrival, motor variant, and a cranial nerve deficit were significantly associated with severe GBS.
Waalgard et al. found that an age <40 years, a high rating of Medical Research Council sum score (MRCSS), mild GBS (Hughes 0–2) at admission, and a previous URI were associated with a good outcome (ability to walk independently at 3 and 6 months). Moreover, higher age, previous diarrhea, and low MRCSS at hospital admission and at one week were independently associated with poor outcome (unable to walk at 3 and 6 months). In our study, a severe Hughes score (3–5) at hospital admission and a severe Hughes score (3–5) at hospital discharge were also associated with poor outcomes at 3 and 6 months. In this study, we were able to identify early gait complaints as a predictor of a good outcome at discharge; this may be due to a mild and limited presentation of the disease with early nonbulbar motor involvement. This finding should be verified with new prospective controlled studies to establish its real statistical value. A more recent European study about predictors of poor outcome at 6 months reported a significant association between high age, previous diarrhea, and a severe GBS disability score.
The strength of the present study was its design as a consecutive prospective study and its systematized and accurate information obtained by only one tertiary hospital center. The main weakness was the loss of patient follow-up at 3 and 6 months. Patients with GBS and equivocal patterns did not have new electrophysiological studies to see if the pattern changed or was eventually defined at follow-up. Also, in-hospital complications were not specified, and these could affect the secondary analysis of mortality or poor outcome at hospital discharge and follow-up. We consider that performing this study in a tertiary referral hospital with a higher number of severe GBS severe cases is one of the main limitations. It is necessary to conduct a secondary replica in primary care hospitals to obtain more representative results nationwide.
Axonal pattern, motor involvement, and the descendent pattern of weakness at presentation were the main clinical findings in our patients with GBS. A high Hughes score at hospital admission is a strong predictor for an adverse outcome at hospital discharge and a short-term outcome, independent of treatment type or in-hospital area management. GBS carries considerable mortality in Mexico, and more studies are needed to address this concern.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3], [Table 4]