A retrospective observational study of clinical and electrophysiological types of Guillain-Barre syndrome from Delhi

ABSTRACT

Objectives:

To understand the demographic characteristics (age, sex and comorbidities), distribution of electrodiagnostic subtypes, and severity and prognosis of Guillain-Barre syndrome (GBS) variants.

Materials and Methods:

This 5-year retrospective study was conducted between January 2018 and December 2023. The patients (n = 137) were diagnosed using the NINDS criteria. Severity and prognosis were assessed using the Hughes disability score. Electrodiagnostic variants of GBS were labelled using a nerve conduction study.

Results:

The patient group showed male predominance with an increased incidence with age. Antecedent infections were observed in 52 (38%) patients. Motor weakness, areflexia, and cranial nerve involvement (VII,IX,X,XI) were the most common clinical manifestations. The electrodiagnostic profiling included 96 (70%) patients with acute demyelinating polyradiculoneuropathy (AIDP), 25 (18.24%) patients of axonal forms and 16 (11%) patients of Bickerstaff Brainstem Encephalitis, Miller Fisher syndrome and paraparetic variants. Albumin cytological dissociation was documented in 61 (70%) patients out of 86 patients, which was more marked in AIDP than in other variants. The severity of GBS, using the Hughes disability score, was higher in patients with underlying diabetes mellitus. Recurrent GBS was observed in three (2%) and mortality rate was 2.91%.

Conclusions:

Our study showed a higher incidence of demyelination than that of axonal variants. There was more variation in albumin cytological dissociation in demyelinating subtypes than in the other variants. GBS severity was significantly higher in the patients with diabetes mellitus.

Introduction

Guillain-Barre syndrome (GBS) is an immune-mediated, inflammatory disease of the peripheral nervous system (PNS) and is one of the common causes of acute flaccid paralysis. It is a monophasic illness that typically presents with relatively symmetrical weakness of the limbs, which rapidly progresses over time. It may or may not be preceded by sensory symptoms (tingling and numbness) in the limbs, even days before the onset of motor weakness. The annual incidence of the disease is reported to be 1–2/100,000 per year, with male predominance (1.5:1). Children are reported to be less affected (0.34–1.34 per 100,000 per year) than adults.[1] Epidemiological studies of GBS are not easy to interpret because of their associated comorbidities, giving it a wide array of presentations. Clinical presentation may be masked by underlying systemic diseases such as diabetes or carcinoma; therefore, the pure form may be difficult to elicit.

Method

This retrospective observational study included patients of all age groups from January 2018 to December 2023 in New Delhi, India. Data were collected from the medical records of hospitals using the ICD 10 classification code for Acute Inflammatory Demyelinating Polyneuropathy (AIDP), polyneuropathy and GBS. The patient was diagnosed using the National Institute of Neurological Disease and Stroke (NINDS) diagnostic criteria for GBS.[2] Hughes disability score (HDS) was used as a severity and prognostic score for the quantitative assessment of GBS patients.[3] Ethical clearance was obtained from the institutional ethical committee before data collection. This study meets all Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [Figure 1]

Figure 1.

Consort diagram depicting data acquisition

Data entry and statistical analysis

The collected data were transformed into variables, coded and entered in Microsoft Excel. Data were analysed and statistically evaluated using the SPSS-PC-25.

The normal distribution of the different parameters was assessed using the Shapiro-Wilk normality test. Quantitative data were expressed as mean ± standard deviation or median with interquartile range and depended on normality. The difference between the means of the two groups was compared using the Mann–Whitney U test, while for more than two groups, the Kruskal–Wallis H test was used. Qualitative data were expressed as frequencies and percentages and statistical differences between proportions were assessed using the Chi-square test or Fisher’s exact test. The P value was set as < 0.05.

Result

A total of 137 patients were studied, including 47 patients admitted during the COVID-19 era.

The mean age of the patient was 49.95 ± 18.95. Male predominance was seen in all age groups except in 41–60 years, where there was female predominance (11.3%) [Table 1 and Figure 2].

Table 1.

Age and gender-wise distribution of GBS study subjects (n=137)

	Female	Male
Up to 20 years	3 (5.6%)	9 (10.8%)
21–40 years	10 (18.5%)	18 (21.7%)
41–60 years	23 (42.6%)	26 (31.3%)
>60 years	18 (33.3%)	30 (36.1%)
Total	54 (39.4%)	83 (60.6%)

Mean age was 49.95±18.95 years.

Figure 2.

Age wise distribution of GBS study subjects

Comorbidities: Diabetes Mellitus (DM) was present in 33.6%, hypertension in 39.4%, and other comorbidities (seizure, coronary artery disease, metastasis, and hypothyroidism) were observed in 26.3% of our cohort [Table 2].

Table 2.

Comorbidities in study subjects (n=137)

Comorbidities	Number	%
DM	46	33.6
Hypertension	54	39.4
Others	36	26.3

Antecedent infection (upper respiratory tract infection, acute gastroenteritis, fever, and surgery) was observed in 38% of patients with a mean of 30 ± 4 days [Table 3].

Table 3.

Antecedent infection in study subjects (n=137)

Antecedent infection	Number	%
No	85	62.0
Yes	52	38.0

The average hospitalisation duration for the demyelinating variant was 9.59 days, whereas that for the axonal variants was 19.9 days.

Of the 137 patients, 96 had AIDP and 25 were axonal variants (AMAN and AMSAN) [Figure 3].

Figure 3.

GBS VARIANTS. AIDP: Acute inflammatory demyelinating polyneuropathy, PCB: Pharyngo cervical brachial, AMAN: Acute motor axonal neuropathy, BBE: Bickerstaff brainstem encephalitis, MFS: Miller fisher syndrome, AMSAN: Acute motor sensory axonal neuropathy

Clinical variation of GBS

1. On Admission: HDS-2 was observed in three (3.1%) AIDP patients. HDS-3 was observed in 57 (59.4%) AIDP and 10 (40%) axonal variants (AMAN and AMSAN) patients. HDS-4 was observed in 27 (28.1%) AIDP and 12 (48%) axonal variants. HDS-5 was observed in 9 (9.4%) AIDP and 3 (12%) axonal variants. P >0.05, statistically insignificant between the GBS subtypes [Table 4].

2. On discharge- HDS-2 was observed in five (5.2%) AIDP patients. HDS-3 was observed in 58 (60.4%) AIDP and 17 (68%) axonal variants (AMAN and AMSAN) patients. HDS-4 was observed in 23 (24%) AIDP and 5 (20%) axonal variants. HDS-5 was observed in seven (7.3%) AIDP and 2 (8%) axonal variants. HDS-6 was observed in 3 (3.1%) in AIDP and 1 (94%) in axonal variants. P >0.05, statistically insignificant between GBS subtypes [Table 5].

3. The mean number of hospitalisation days for all GBS cases was not significantly different (P > 0.05). Albumino cytological dissociation (ACD) was significantly higher in the AIDP variants (73.5 mg/dL) than in the axonal variant, AMSAN (36.5 mg/dL), P = 0.001 [Table 6].

4. Cranial nerve involvement was seen in 6 patients (14.63%) presented with facial diplegia, and 27 patients (65.85%) had bulbar palsy, of which 7 patients (25.92%) were mechanically ventilated, which was statistically significant [Table 7].

Table 4.

Association of GBS variants with severity at admission in study subjects (n=137)

HDS at admission	AIDP (n=96)	AMAN/ASMAN (n=25)	Others (n=16)
2	3 (3.1%)	0	2
3	57 (59.4%)	10 (40%)	9
4	27 (28.1%)	12 (48%)	5
5	9 (9.4%)	3 (12%)	0

P=0.12

Table 5.

Association of GBS variants with severity at discharge in study subjects (n=137)

HDS at discharge	AIDP (n=96)	AMAN/ASMAN (n=25)	Others (n=16)
2	5 (5.2%)	0	3
3	58 (60.4%)	17 (68%)	10
4	23 (24%)	5 (20%)	1
5	7 (7.3%)	2 (8%)	1
6	3 (3.1%)	1 (4%)	1

P=0.36

Table 6.

Association of GBS variants with no of hospitalisation days and ACD in study subjects (n=137)

	AIDP (n=96)	AMAN (n=19)	AMSAN (n=6)	Others (n=16)	P
No of hospitalisation days	7 (6–10)	7 (6–13)	7.5 (5.75–10.75)	9 (6.25–15.75)	0.54
ACD	73.5 (51–147)	68 (47.62–84.6)	36.5 (21.25–53.25)	25.65 (22.25–43.75)	<0.001

Table 7.

Association of cranial nerve involvement with a need for mechanical ventilation in study subjects (n=137)

Mechanical ventilation	No cranial nerve involvement	Cranial nerve involved	P
Not needed	100	13	0.02
Needed	17	7

Association of GBS variants with number of hospitalisation days and ACD

The mean CSF protein levels in our study were 131.76 mg/dL in AIDP and 105.92 mg/dL in axonal variants of GBS. There was no significant difference in ACD values between the diabetic and non-diabetic groups [Table 8].

Table 8.

Association of DM with ACD in study subjects

	DM −nt	DM +nt	P
ACD	65 (38–119)	64 (49.5–127)	0.56

Discussion

This was a retrospective observational study on the incidence of GBS over five years in North India. Our study showed a higher incidence of AIDP (70% of patients) as compared to axonal (18.24% of patients) variants. As per the literature, it varies around the world, with AIDP (60–80% of patients) predominantly in North America and Europe and AMAN (30–65% of patients) in Asia, Central America and South America.[4,5,6,7,8] This variation can be attributed to genetic polymorphisms due to geographical diversity, differences in genetic backgrounds and environmental exposures. Nutritional status and socio-economic status corresponding to hygienic conditions, which correspond to an increased incidence of axonal variants due to common bacterial infections (Campylobacter jejuni), are profound in low socio-economic countries.[4,9]

We found a high incidence of disease, 69.4%, in the elderly age group (>60 years of age), possibly due to lower immunity with ageing. The incidence increased by approximately 20–30% with an increase in age. The lower incidence in children could be attributed to the lower number of children reported to the hospital. This could be due to a misdiagnosis or the self-remitting nature of the disease. There were 11.3% more female patients seen in the elderly age group (>60 years) than in younger age groups with male predominance. GBS has a bimodal distribution of age in GBS, with a 20% increase chances with increase in age. Delannoy et al., in a French population (n = 9391), reported a childhood incidence peak around the age of 2 years, which is supported by the reports from China by Tang et al.[10,11] There is disparity in gender predominance, in contrast to earlier reports of male predisposition in GBS. McCombe et al. reported a female preponderance, whereas González-Suárez et al. showed no sex difference in GBS.[12,13] GBS is an autoimmune disease, apart from Goodpasture’s syndrome, with a predilection for males.[1,14]

There have been conflicting reports mentioning an increase in the incidence of GBS with changing seasons, which was not supported by our study.[1,15]

Our study showed 52 (37.95%) patients had antecedent infections or triggering factors that contributed to GBS. These were documented as acute gastroenteritis, upper respiratory tract infection, transient fever and abdominal and cardiac surgeries. This number can also be attributed to the inclusion of the 2-year pandemic era in our study. Antecedent infections (Campylobacter Jejuni, Cytomegalo Virus, and influenza virus), Zika Virus and COVID-19 epidemics and pandemics, and influenza vaccination are known to trigger GBS. Patients with a history of surgery, especially neurosurgery, gastrointestinal surgery, vascular surgery, orthopaedic surgery and cardiovascular surgery, are 5.8 times more likely to develop GBS. Shahrizaila et al. reported that 76% of patients had triggering factors 4–6 weeks before the onset of GBS. Immune checkpoint inhibitors, salazopyrine drugs and inflammatory bowel disease have also been associated with GBS.[6,16,17,18,19]

In our study, out of 137 patients, 24 patients (25%) required mechanical ventilation due to respiratory distress, 20 patients (14.63%) had facial diplegia and 21 patients (15%) manifested dysautonomia. One patient in our study presented with unilateral lateral rectus nerve palsy with progressive weakness in all four limbs.

The diagnostic criteria for GBS were introduced by the NINDS in 1978 and include a symmetrical progression of weakness over 4 weeks, areflexia or hyporeflexia, mild sensory symptoms, ACD and nerve conduction studies as supportive parameters. It is a clinical diagnosis; therefore, ruling out differential diagnosis is important.

Among the low-incidence variants of GBS are the pure sensory variant and pan dysautonomia (tachycardia, hypertension, gastrointestinal dysfunction, and bladder dysfunction), which can easily mimic other diseases. Disease progression may be rapid with nadir (lowest MRC score or highest HDS) achieved by 2–4 weeks (80% of patients in 2 weeks and 97% of patients in 4 weeks). The clinical presentation can vary from weakness of the limbs with increasing severity owing to the involvement of cranial nerves (facial diplegia, dysphagia and bulbar weakness), respiratory muscles (respiratory distress) and brain parenchyma (encephalitis).[4,20]

Our study included 16 patients with atypical variants of GBS including Bickerstaff brainstem encephalitis (BBE) (2 patients), Miller Fisher syndrome (MFS) (6 patients), polycranialis neuritis (2 patients), bulbar variant (1 patient), paraplegic variant (two patients) and pharyngocervicobrachial variant (PCB) (3 patients). MFS and BBE form part of the spectrum of the anti-GQ1b antibody syndrome. These show loss of Hoffman reflex and detection of 1 Hz sway in body sway analysis, indicating the involvement of 1a fibres, found to be affected in 70% of the patients. There is an overlap between the CNS and PNS manifestations of clinical symptoms in these GBS variants.[21,22]

Two patients in our study, the BBE and AMAN variants, showed hyperreflexia and normal reflexes, respectively, on admission. Literature shows 9% with normal myotatic reflexes in upper limb weakness and 2% may have normal myotatic reflexes in the lower limb despite weakness. Gross asymmetry is an unusual feature of GBS, which was observed in one patient with an axonal variant and very high CSF protein levels (980 mg/dL).[23]

A nerve Conduction Study (NCS) was conducted 7–10 days after the onset of symptoms in our study in four motor nerves and two sensory nerves of both the upper and lower limbs. The H- reflex was absent in all demyelinating variants of GBS. Among the atypical variants (16 patients) of GBS, 9 had normal NCS. Three patients underwent conduction blocks without temporal dispersion (1 in the upper limb nerve and 2 in the lower limb nerves). Serial conductions were not performed to investigate reversible conduction failure, which is a limitation of this study.[24] NCS is preferred two weeks after symptom onset. The earliest findings could be a non-recordable Hoffman reflex and an absent/prolonged F wave. A better diagnostic yield was obtained when four motor nerves and three sensory nerves were studied. Uncini and Kuwabara,[24] with serial conduction studies, reported reversible conduction failure (RCF) with a transient conduction block without temporal dispersion and normal amplitude which recovered rapidly with the disease course. This variant is termed axonal motor neuropathy with conduction block (AMNCB), a form of nodopathy in which IgG4 antibodies are formed against gangliosides.

Our study included two years of the COVID-19 era in which lumbar puncture (LP) was not feasible for diagnosis. Of the 137 patients, LP was possible in 86. Of these, 25 were acellular with normal protein levels. Oedema in the nerve root is probably caused by cyto-albumin dissociation because proteins with at least 70 kDa albumin cannot dissociate toward the blood compartment. Ideally, CSF examination is performed 7–10 days after disease onset. CSF protein levels were significantly higher in demyelinating variants than in axonal variants, with poor outcomes or severity. ACD is found in 50% of patients when performed within 3 days and in 80% of patients when conducted in 7–10 days.[18] Normal CSF proteins do not rule out GBS, but raised cells > 50 cells/μL can raise the suspicion of other mimics like leptomeningeal malignancy, infections or inflammation of spinal roots. An increase in CSF proteins (normal range 15–45 mg/dL) is attributed to myelin damage and antibody deposition which makes the blood nerve barrier permeable. Intravenous immunoglobulin (IVIG) can further increase proteins and cells in the CSF due to transudation or meningitis caused by breakdown of the blood brain barrier; therefore, lumbar puncture during its administration or repeated LP is avoided.[4,19,23]

Typically, GBS is a monophasic illness, but there is approximately 5% rate of recurrence which was seen in our study as well. Amongst these, 2% of patients had predominant demyelinating variants and younger mean age. According to the literature, the interval of recurrence can be anywhere between 2 months and 39 years.[18]

Most of the patients (60–80% of patients) in our study recovered within 6 months. Sensory symptoms such as tingling and paraesthesia were observed in 52% of the patients. Literature reviews indicate that painful paresthesia, lumbago, fatigue, and meningism may persist even after 1–2 years of the disease remission in 54–89% of patients. These symptoms precede motor symptoms in one-third of patients.[5,24,25,26] Poor predictors of prognosis, as seen in our study, are age, dysautonomia and axonal variants.[6]

In our cohort, the mortality rate was 2.91%, which was lower than in high-income countries (5%). This could be attributed to timely diagnosis and better medical management in our tertiary care centre. Literature reports mortality in low-income countries to be higher (17%), possibly due to limited resources and/or treatment options, with 4% mortality during the recovery period.[2,6,27]

GBS in our 137 patients were managed with IVIG, PLEX or both or immunosuppression with rituximab injection. IVIG is the preferred choice of treatment for most autoimmune diseases, including GBS, owing to the ease of administration and minimal logistics involved compared to PLEX. Seven patients were administered PLEX due to the rapid progression of the disease, causing respiratory distress upon admission. Rituximab injection (for nodopathy) was administered to two patients after IVIG administration because of a poor response to immunomodulatory therapy two months after disease onset. Two patients received IVIG after PLEX as combination therapy for better and faster recovery. The Cochrane review, for the management of GBS, gives equal weight to both IVIG and PLEX to achieve an early plateau. In contrast, few studies have supported the idea of no management strategy because of the self-remitting nature of the disease. Approximately 40% of patients fail to respond to immunomodulatory therapy in the first 4 weeks of treatment.[2]

The clinicians should be aware about the treatment approach (PLEX) where the progression is fast or bulbar and respiratory involvement is seen within seven days of symptom onset. The limitations of PLEX include sepsis, hypotension and dysautonomia, and they should always be considered. Despite the advent of newer technologies and therapies, there is a lack of specificity for the drug targeting the causative agent for GBS. Therefore, the choice of drugs is not different despite the differences in the mechanism of pathogenesis (axonal or demyelinating). This is an area implored for better prognosis and early patient recovery.[27]

Limitation

In a retrospective study, such as ours, where data were retrieved from the medical record room of hospitals, chances of incomplete information or details were always present. GBS is a clinical diagnosis that depends on the subjective assessment and history of symptom onset. The examination performed might have been biased, and therefore, interpreted subjectively by different clinicians at various points of time during the disease course. The imprecise onset of symptoms, which is subjective, can also lead to the diagnosis of chronic inflammatory demyelinating polyneuropathy such as GBS. The diagnostic criteria can be used to interpret all variants of GBS used in this study, such as MFS and BBE. A serial NCS, for RCF was not performed to analyse or diagnose nodopathy (acute motor axonal neuropathy with conduction block).

Conclusion

This study demonstrated increased levels of CSF proteins (albumin) in demyelinating variants compared with other variants of GBS. The mortality rate is less than 5%, credited to faster diagnosis and timely medical management, compared to the reported mortality rate of lower-income countries. The severity of the disease is higher in GBS patients with underlying diabetes mellitus and cranial nerve involvement. There is still no consensus on the treatment choice for GBS variants to minimise morbidity and improve prognosis. GBS is a monophasic disease with a chance of recurrence, which might be severe on presentation and could be resistant to the indicated management regimen. Therefore, it is imperative to study and understand disease mechanisms and identify specific biomarkers for prognostication and treatment efficacy.

Consent for publication

Yes.

Availability of supporting data

Yes, the data is available on request.

Authors contribution

All authors confirm the sole responsibility for study conception, design, analysis and manuscript preparations.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.