Electrical vagus nerve stimulation is a promising approach to reducing pulmonary complications after an esophagectomy: an experimental rodent model

Abstract

After esophagectomy, an imbalanced inflammatory response increases the risk of postoperative morbidity. The vagus nerve modulates local and systemic inflammatory responses, but its pulmonary branches are transected during esophagectomy as part of the oncological resection, which may account for the high incidence of postoperative (pulmonary) complications. This study investigated the effect of electrical vagus nerve stimulation (VNS) on lipopolysaccharide (LPS)-induced lung injury in rats. Rats (n = 60) were randomly assigned to a non-vagotomy or cervical vagotomy group, with VNS or without (NOSTIM). There were four non-vagotomy groups: NOSTIM and bilateral VNS with 100, 50, or 10 µA. The four vagotomy groups were NOSTIM and VNS with fixed amplitude (50 µA) bilaterally before (VNS-50-before) or after bilateral vagotomy (VNS-50-after), or unilaterally (left) before ipsilateral vagotomy (VNS-50-unilaterally). LPS was administered intratracheally after surgery. Pulmonary function, pro-inflammatory cytokines in serum, broncho-alveolar lavage fluid (BALF), and histopathological lung injury (LIS) were assessed 180 min post-procedure. In non-vagotomized rats, neutrophil influx in BALF following intra-tracheal LPS (mean 30 [± 23]; P = 0.075) and LIS (mean 0.342 [± 0.067]; P = 0.142) were similar after VNS-100, compared with NOSTIM. VNS-50 reduced neutrophil influx (23 [± 19]; P = 0.024) and LIS (0.316 [± 0.093]; P = 0.043). VNS-10 reduced neutrophil influx (15 [± 6]; P = 0.009), while LIS (0.331 [± 0.053]; P = 0.088) was similar. In vagotomized rats, neutrophil influx (52 [± 37]; P = 0.818) and LIS (0.407 [SD ± 0.037]; P = 0.895) in VNS-50-before were similar compared with NOSTIM, as well as in VNS-50-after (neutrophils 30 [± 26]; P = 0.090 and LIS 0.344 [± 0.053]; P = 0.073). In contrast, VNS-50-unilaterally reduced neutrophil influx (26 [± 10]; P = 0.050) and LIS (0.296 [± 0.065]; P = 0.005). Systemic levels of cytokines TNF-α and IL-6 were undetectable in all groups. Pulmonary function was not statistically significantly affected. In conclusion, VNS limited influx of neutrophils in lungs in non-vagotomized rats and may attenuate LIS. Unilateral VNS attenuated lung injury even after ipsilateral vagotomy. This effect was absent for bilateral VNS before and after bilateral vagotomy. It is suggested that the effect of VNS is dependent on (partially) intact vagus nerves and that the level of the vagotomy during esophagectomy may influence postoperative pulmonary outcomes.

Keywords: Esophagectomy; Inflammation; Pulmonary complications; Vagotomy; Vagus nerve; Vagus nerve stimulation.

Fig. 1

Schematic overview of the experiment. The specific surgical procedure—sham procedure versus vagotomy—and method of electrical stimulation are described in detail in the methods. Abbreviations: LPS, lipopolysaccharide; BALF, broncho-alveolar lavage fluid; SHAM, sham procedure without vagotomy (non-vagotomy); VGX, cervical vagotomy; NOSTIM, group without VNS; VNS, electrical vagus nerve stimulation

Fig. 2

Position of rats during procedure. With the rats in the supine position, the electrode was brought gently towards the vagus nerves, after which the vagus nerves were stimulated three times in 10 min using fixed parameters: constant-current, biphasic, and charge-balanced square wave pulses (10, 50, or 100 µA/phase, 2 ms/phase at 5 Hz) for 1 min each depending on the groups as described in the methods

Fig. 3

Number of inflammatory cells in BALF (× 10⁴) in non-vagotomized rats. A Total cell count, B macrophages, and C neutrophils. Bilateral VNS before LPS (intratracheally) significantly reduced the influx of neutrophils after 50 and 10 µA, but not 100 µA compared to NOSTIM. Cell count of macrophages was also lower after 10 µA, but not 50 or 100 µA. Values are means with standard deviation. A P-value of < 0.05 after post-hoc Dunn’s test was considered statistically significant. Abbreviations: BALF, broncho-alveolar lavage fluid; LPS, lipopolysaccharide; NOSTIM, group without VNS; VNS, vagus nerve stimulation

Fig. 4

Histopathological lung injury score (LIS) in non-vagotomized rats. Histological sections of the lungs (H&E stained) at × 400 magnification (scale bar represents 50 µm). The NOSTIM group (A) showed some thickening of alveolar septa (marked by arrowheads) and neutrophil infiltration in the interstitial and alveolar spaces (marked by arrows). Alveolar mononuclear cell infiltration (e.g., plasma cells and macrophages) is observed (marked by M). LIS was not markedly affected by VNS with 100 µA (B). LIS was significantly reduced after VNS with 50 µA (C), whereas after 10 µA (D), this finding did not reach statistical significance. Note that the alveolar walls are thin and the alveoli contain occasional alveolar mononuclear cells (C and D). Values are means with standard deviation. A P-value of < 0.05 after post-hoc Dunn’s test was considered statistically significant. Abbreviations: NOSTIM, group without VNS; VNS, vagus nerve stimulation. N = 5 in all groups

Fig. 5

Pulmonary function in non-vagotomized rats. There were no statistically significant changes in A lung resistance or B dynamic compliance between the NOSTIM and VNS groups. Values are medians (interquartile range is not shown for clarity). A P-value of < 0.05 was considered statistically significant. P-values shown are Bonferroni-adjusted (multiplied by 3) as compared to NOSTIM. Abbreviations: NOSTIM, group without VNS; VNS, vagus nerve stimulation

Fig. 6

Number of inflammatory cells in BALF (× 10⁴) in the vagotomy group. A Total cell count, B macrophages, and C neutrophils. Bilateral VNS-50-before (50µA(B)) bilateral vagotomy and LPS (intratracheally) reduced the number of macrophages, but not neutrophils compared to NOSTIM. Bilateral VNS-50-after (50µA(ST); i.e., efferent vagal stumps) bilateral vagotomy reduced the number of neutrophils, but this did not reach statistical significance. VNS-50-unilaterally (50µA(L)) before ipsilateral (left) vagotomy reduced the influx of neutrophils and macrophages. Values are means with standard deviation. A P-value of < 0.05 after post-hoc Dunn’s test was considered statistically significant. Abbreviations: BALF, broncho-alveolar lavage fluid; LPS, lipopolysaccharide; NOSTIM, group without VNS; VNS, electrical vagus nerve stimulation

Fig. 7

Histopathological lung injury score (LIS) in the vagotomy group. Histological sections of the lungs (H&E stained) at × 400 magnification (scale bar represents 50 µm). The NOSTIM group (A) showed some thickening of alveolar septa (marked by arrowheads) and neutrophil infiltration in the interstitial and alveolar spaces (marked by arrows). Alveolar mononuclear cell infiltration (e.g., plasma cells and macrophages) is observed (marked by M). LIS was not affected by VNS-50-before (50µA(B)) bilateral vagotomy (B). VNS-50-after (50µA(ST); i.e., efferent stumps) bilateral vagotomy reduced LIS, but not statistically significantly (C). VNS-50-unilaterally (50µA(L)) before ipsilateral (left) vagotomy significantly reduced LIS (D). Note that the alveolar walls are thin and the alveoli contain occasional alveolar mononuclear cells (C and D). Values are means with standard deviation. A P-value of < 0.05 after post-hoc Dunn’s test was considered statistically significant. Abbreviations: LPS, lipopolysaccharide; NOSTIM, group without VNS; VNS, electrical vagus nerve stimulation. N = 5 in all groups

Fig. 8

Pulmonary function in the vagotomy group. There were no statistically significant changes in (A) lung resistance or (B) dynamic compliance between the NOSTIM and VNS groups. Values are medians (interquartile range is not shown for clarity). A P-value of < 0.05 was considered statistically significant. P-values shown are Bonferroni-adjusted (multiplied by 3) as compared to NOSTIM. Abbreviations: NOSTIM, group without VNS; VNS, electrical vagus nerve stimulation; 50µA(B), VNS-50-before bilateral vagotomy; 50µA(ST), VNS-50-after bilateral vagotomy (i.e., efferent stumps); 50µA(L), VNS-50-unilaterally (left) before ipsilateral vagotomy