Small Fiber Neuropathy in Pulmonary Sarcoidosis and Tuberculosis: Clinical and Histological Correlates

Sarcoidosis (SC) is the granulomatous disease of an unknown origin, where the differential diagnosis with tuberculosis (TB) is challenging and vital for patients’ prognosis. The common neurological complication in SC is a small ber neuropathy (SFN), that is considered to be the result of the chronic inammation, and remains signicantly understudied. There is no reliable data, whether such complication is observed in TB patients, where the systemic inammation is also described. Aim: To identify the clinical and histological correlates of the small ber neuropathy in sarcoidosis and tuberculosis patients. one SFN. included dysfunctions of the musculoskeletal and A negative, statistically signicant correlation between the nerve ber (IEND) and SFN-SL score revealed in both groups (Spearman r -0.3508, p 0.0102, and r -0.7382, p 0.0064, respectively). the density of small in the patients with was lower, compared to the with test, symptoms of the small ber neuropathy were subsequently less prominent, which may represent the difference between the autoimmune and bacterial inammation. The validated questionnaires and histologic verication of the diagnosis help to establish the severity of neuropathy of small bers, to determine the prognosis, to plan the treatment strategн, and also may allude the possibility for the additional criteria of differential diagnosis between two diseases.


Introduction
In patients with sarcoidosis, the development of systemic in ammation and internal organ dysfunction are observed, which signi cantly reduces the quality of life and worsens the prognosis of the patients. One of the most common complication is considered to be a small ber neuropathy (SFN), that remains signi cantly understudied (1)(2)(3)(4).
The prevalence of sarcoidosis varies throughout the world. In Japan there is 1 case per 100,000 people, while in Scandinavian countries the prevalence of sarcoidosis is as much as 63 cases per 100,000 people, In Russian Federation sarcoidosis is described with the prevalence from 22 to 47 cases per 100,000 people, depending on the region (12)(13)(14)(15). The prevalence of SFN may also vary because it presents not only with neuropathic pains and paresthesias, but also with various symptoms of autonomic dysfunction, which may not be recognized as a neurologic complication (16)(17)(18)(19)(20).
The development of SFN is considered to be the result of a cytokine-mediated in ammation, which is typical for various autoimmune diseases, including sarcoidosis (5)(6)(7)(8). Small nerve ber damage is also observed in systemic lupus erythematosus, Sjogren's syndrome, and bromyalgia (9,10). Considering the signi cant role of the genetic predisposition and the possible provocative role of exogenous triggers in the development of this complication, SFN in patients with sarcoidosis can be considered as a part of the autoimmune/in ammatory syndrome induced by adjuvants (ASIA) (11). Several cases of the proven SFN were also observed in patients, suffering from the bacterial in ammation, e.g., Lime disease and leprosy (1)(2)(3)(4). To our best knowledge, there is no studies on the detecting the reduction of the small nerve bers in patients with tuberculosis, though this data may contribute to the better understanding of the pathogenesis of different types of in ammation and also allude the new marker of the differential diagnosis between tuberculosis and sarcoidosis.
Currently, there are no generally accepted criteria for the diagnosis of SFN. The presence of a small nerve bers dysfunction in a patient usually is based mainly on clinical criteria, such as neurological examination and validated scales (for example, small ber neuropathy screening list) (19,(24)(25)(26). In addition, electroneuromyography can be performed mainly to exclude damage to large nerve bers. The "gold standard" for diagnostics is immuno uorescence or immunohistochemistry of skin biopsy with the calculation of the density of intraepidermal nerve bers. This technique requires special training and equipment, and is signi cantly time consuming (27)(28)(29)(30)(31)(32)(33). Preliminary clinical diagnosis of SFN in patients with sarcoidosis is important due to the low awareness of healthcare practitioners about this complication and the need for a quick assessment of neuropathy signs for the further skin biopsy performance.

Aim Of The Study
To identify the clinical and histological patterns and correlates of the small ber neuropathy in lung sarcoidosis and lung tuberculosis patients for differential diagnostic. The study entailed 119 patients with lung sarcoidosis and lung tuberculosis, and healthy subjects. Among them were 42 men and 56 women, average age was 38.4 ± 7.2 years. The rst group consisted of patients with lung sarcoidosis (n = 25, average age 33.4 ± 8.5 years), the second group -patients with veri ed lung tuberculosis before initiating speci c therapy (n = 21, average age 36.6 ± 9.3 years), the third group included healthy subjects (n = 25, average age 43.2 ± 11.7 years). There were no statistically signi cant differences in gender and age among the patient groups.

Materials And Methods
The inclusion criteria were age from 18 to 65 years and signing informed consent to participate in the study. The patients from the group 1 were also diagnosed with sarcoidosis, stage I-II, for group 2 -focal and disseminated lung tuberculosis in the pre-treatment period.
Exclusion criteria were: A hormonal therapy, the presence of Löfgren's syndrome and a chronic course of the disease (for patients with sarcoidosis), the presence of other infectious diseases (HIV, hepatitis C), the history of cancer and their treatment using chemotherapy, the presence of other diseases (diabetes, hypothyroidism, renal failure, vitamin de ciency or overdose), medications (metronidazole, nitrofurantoin, linezolid, ecainide, statins), as well as the medical history of autoimmune diseases.
According to the design of the study, patients with sarcoidosis and pulmonary tuberculosis underwent a standard examination using X-ray, morphological, bacteriological and molecular genetic methods. A neurological examination with an assessment of the super cial and deep pain sensitivity, muscle strength and muscle-tonic re exes was performed. For clinical veri cation of the SFN, the validated questionnaire for the detection of small ber neuropathy was used (Small ber neuropathy screening list, SFN-SL, (19)). The questionnaire consists of 2 parts and 21 questions, from 0 to 4 points each, evaluating both the frequency of development of symptoms and their intensity. A moderate likelihood of neuropathy is established when the diagnostic threshold is reached at 22 points, high -at 48 points. With scores less than 11, sensitivity is 100%, speci city is 31.0%. With scores of more than 48 -sensitivity -19.0%, speci city -100%.
In 13 patients with lung sarcoidosis and 10 patients with lung tuberculosis, a punch biopsy of the skin was performed (10 cm proximal to the external malleolus) followed by the xation of the specimen in Zamboni solution and the performance of enzyme immunoassay with primary PGP 9.5 antibodies (Abcam) and secondary AlexaFluor goat-anti rabbit antibodies 488 (Abcam). The results were compared with the normal values obtained in the worldwide normative reference study [23].
Statistical analysis was performed using the Statistica 8.0 software package. The distribution of patients into groups was carried out in accordance with the presence of veri ed diagnoses of lung sarcoidosis, lung tuberculosis or in the absence of veri ed somatic diseases (for healthy subjects).
All data collected during the study were analyzed using methods of descriptive statistics. The data with the normal distribution are presented in the form of the "mean ± standard deviation" formula. Data that were not normally distributed are presented in the form of the "median (interquartile range 25 -75 quartiles)", where the con dence interval was also calculated. Normality testing was performed using the Shapiro-Wilk test.
All data were analyzed using methods of parametric and nonparametric statistics, Chi-square test and statistical comparison methods for two (Mann-Whitney U-test) and three (Kruskal-Wallis test) groups were performed, correlations were carried out using the Spearman coe cient.
Differences or association rates were considered statistically signi cant at a p<0.05.
Statistically signi cant differences were found in the results of patients with sarcoidosis and tuberculosis (p = 0.0012, the Mann-Whitney U-test), in patients with sarcoidosis and healthy subjects (p <0.0001), as well as in patients with tuberculosis and healthy subjects (p = 0.03). Thus, in the rst group, a higher average SFN-SL score was observed than in the second group, while the results of the 1 and 2 groups exceeded those of healthy individuals.
The main clinical symptoms of the SFN were: pain, changes in body temperature, impaired motility of the gastrointestinal tract and urinary system, heart palpitations ( Table 2).
In patients with sarcoidosis, the most frequent clinical symptoms include impaired cardiovascular regulation (arrhythmias, orthostatic hypotension), pain in the chest or in the extremities, and blurred vision. It is important to note that in 76% of patients with sarcoidosis, clinical symptoms had a severity of no more than 1 point on the SFN-SL scale and, in most cases, did not bother patients or lead to a decrease in the quality of life. Among patients with tuberculosis, in general, fewer clinical manifestations were noted, the main of which were dysfunction of the gastrointestinal tract, muscle cramps and chest pain ( Table 3).
Twenty three biopsy specimens have been obtained in patients with sarcoidosis and tuberculosis, where the intraepidermal density (IEFD) of small nerve bers was calculated (Fig. 1, 2).
Of the 23 biopsies, 13 were obtained in patients with sarcoidosis, 10 -in patients with tuberculosis ( Table 4). The calculations of intraepidermal density of the small bers was performed in accordance with the values from the worldwide normative reference study [23].
Thus, in all examined patients, the number of small nerve bers was within normal limits, but below average values. There is a tendency to a higher density of small nerve bers in patients with tuberculosis compared with sarcoidosis (Mann-Whitney test, p = 0.0047). A negative, statistically signi cant correlation was revealed between IEND results and SFN-SL score in patients with sarcoidosis (Spearman's nonparametric rank correlation coe cient, r = -0.3508, p = 0.0102).
A negative, statistically signi cant correlation between IEND results and SFN-SL score was observed in patients with p tuberculosis (Spearman's nonparametric rank correlation coe cient, r = -0.7382, p = 0.0064).
Thus, a decrease in the density of small nerve bers in patients with sarcoidosis was noted, compared with the results of patients with tuberculosis (Mann-Whitney test, p = 0.0047). A negative, statistically signi cant correlation between the IEND and SFN-SL score was described in both groups (Spearman coe cient, r = -0.3508, p = 0.0102, and r = -0.7382, p = 0.0064).  Note: p 1 is the difference between the results of groups 1 and 2, the Chi-square test with Yates correction. p 2 -the difference between the results of groups 1 and 3, Chi-square criterion with Yates correction.  Discussion In our study, neurological disorders were described in 60% of patients with sarcoidosis and 19.1% of patients with tuberculosis. Small ber neuropathy can manifest with a wide range of symptoms, including autonomic and sensory dysfunction. The most common clinical manifestations observed in our study in patients with sarcoidosis were impaired cardiovascular regulation (36% of cases), e.g. the development of cardiac arrhythmias and orthostatic hypotension. In 32% of cases, patients noted pain in the chest or limbs, which often accompanied with allodynia, a subjective perception of tactile touch as pain. In some cases, this can result in sleep disturbances or restless legs syndrome, arising from the discomfort sensations of bed linen touching the skin. Another symptom that is often noted by patients with sarcoidosis is blurred vision, described in 20% of cases. A physician needs to clarify whether visual impairments are transient or permanent in nature, for the differential diagnosis of ophthalmic pathology. In the neuropathy of small bers, blurred vision is transient, arising while overworking or physical exertion. It is considered to be associated with the immune-mediated in ammation of the optic tracts. Less typical for patients with sarcoidosis are disorders of the gastrointestinal tract. In 12% of the cases, patients complained of impaired intestinal motility with the development of both diarrhea and constipation, which occurred simultaneously with the onset of sarcoidosis. This also includes subjective complaints of swallowing dysfunction, which is associated both with impaired muscle innervation and with the progression of dryness of the mouth. In the tuberculosis group much less clinical symptoms were observed. The most common symptoms were the dysregulation of the gastrointestinal (9.5%) and urinal tract motility (4.8%) and arthralgia (9.5%). No cardiovascular complaints were observed in this group and the intensity of the symptoms were much less prominent, compared to the sarcoidosis group.
While a negative, statistically signi cant correlation between the IEND and SFN-SL score was described in both groups (Spearman coe cient, r = -0.3508, p = 0.0102, and r = -0.7382, p = 0.0064), a decrease in the density of small nerve bers in patients with pulmonary sarcoidosis was more prominent, compared with the results of patients with pulmonary tuberculosis (Mann-Whitney test, p = 0.0047).
Thus, neuropathy of small bers seems to be a widespread pathology with the development of multiple organ dysfunction. Given the low awareness of both medical specialists and patients about the development of this complication and the di culties in diagnostic, further study of this issue is required. In patients with pulmonary sarcoidosis, small ber neuropathy may develop as a result of systemic immune-mediated in ammation. Wherein, in tuberculosis patients clinical and histological symptoms of the small ber neuropathy were subsequently less prominent, which may represent the difference between the autoimmune and bacterial in ammation and will be useful in differential diagnostic with both diseases. The validated questionnaires and histologic veri cation of the diagnosis help to establish the severity of neuropathy of small bers, to determine the prognosis and to plan the strategy for treatment, and also may allude the possibility for the additional criteria of differential diagnosis between two diseases.

Declarations
This work was supported by the grant from the Government of the Russian Federation (contract No. 14.W03.31.0009 dated February 13, 2017) on the allocation of the grant for the state support of scienti c researchers conducted under the guidance of leading scientists.

Figure 1
Enzyme immunoassay analysis of skin biopsy in a patient with sarcoidosis (male, 34 years old). Small bers stained with PGP antibody 9.5. Arrows indicate the small nerve bers, penetrating in the epithelium.