The present study shows that the elevation of CK levels is seen in GBS. 27% of patients had the elevation of CK levels in the first 4 weeks after onset. In this study, our classification discovers that all GBS patients with elevated CK levels represent a group of AMAN with axonal degeneration or RCF. Moreover, while a limited number of patients underwent serial electrophysiological examinations, serial examinations also confirmed that GBS with elevated CK levels is AMAN with axonal degeneration or RCF, not a group of AIDP. In the initial electrophysiological examinations, some GBS patients with elevated CK levels were classified as the “AIDP pattern” based on Ho’s criteria, which showed prolongation of DMLs and increase in durations. However, they showed small extent of demyelinating features, which is consistent with RCF in the first place.[11,12,14] Notably, in the follow-up examinations, they showed electrophysiological changes consistent with AMAN with RCF, such as rapid normalization of increased duration.[8, 9, 11-14] In other words, none of GBS patients with elevated CK levels showed changes consistent with true AIDP, such as the development of this prolongation of DMLs and increase of durations in. On the other hand, it was seen in some GBS patients with normal CK levels, suggesting that GBS with normal CK levels includes true AIDP.
Several features of GBS patients with elevated CK levels other than electrophysiological findings also confirm that GBS patients with elevated CK levels represent a group of AMAN. In our study, patients with elevated CK levels frequently had antecedent infections and anti-GM1 antibody, rarely had hypoesthesia and cranial nerve involvement, and did not have urinary retention as signs of autonomic failure; all of the features were consistent with AMAN and not true AIDP.[2, 4-6] Conversely, while the most common antecedent infection in AMAN was gastroenteritis, particularly from Campylobacter jejuni, the most common antecedent infection in GBS patients with elevated CK levels in this study was URTI. Subsequently, in GBS patients with elevated CK levels, there were some with AMAN with anti-GM1 antibody following URTI, while anti-GM1 antibody is usually related not with URTI but with gastroenteritis such as C.jejuni infection. AMAN following URTI, including with anti-GM1 antibody, may be a specific feature of GBS patients with elevated CK levels.
To our knowledge, while we found some case reports of GBS patients with elevated CK levels,[1, 30, 31] there is only one report of a series of GBS patients in whom CK levels was investigated. That report showed a 52% incidence of the elevation of CK levels in GBS and pain associated with the elevation of CK levels in GBS patients; however, the authors did not record the disease subgroup of GBS and consequently did not mention the correlation between GBS disease subgroup and the elevation of CK levels. Therefore, our study is the first study of a series of GBS patients in whom correlation between GBS disease subgroup and elevation of CK levels is investigated, demonstrating that GBS patients with elevated CK levels represent a group of AMAN.
Although the mechanism of elevation of CK levels in GBS is still uncertain, several possible mechanisms were proposed in the literature. One possible mechanism is associated with rhabdomyolysis caused by antecedent infection. This mechanism, however, is unlikely because the patients with rhabdomyolysis showed markedly raised CK levels, which were not consistent with our cases with moderate raised CK levels. Another possible mechanism is that rapid denervation due to axonal damage can result in the release of muscle enzymes. It is also proposed that CK elevation could be caused by painful muscle cramp associated with active denervation on the basis of the finding that most of GBS patients with elevated CK levels developed painful sensation. Regrettably, we could not investigate the association between elevation of CK levels and pain, because of retrospective nature of our study. However, the mechanism that denervation could cause CK elevation is consistent with our observation that elevation of CK levels occurred in AMAN with RCF as well as AMAN with axonal degeneration but not in AIDP, because the denervation could occur in AMAN with RCF as well as AMAN with axonal degeneration but not in demyelination alone.[33, 34] The mechanism could explain our observations that elevation of CK began within 2 weeks after onset and that peak of CK followed nadir of disability within 2 weeks, because denervation follows axonal degeneration within at most 2 weeks in proportion to the distance of the muscle from the injury site. The mechanism could also explain that in some of our patients, the elevation of CK levels occurred within a few days, because motor nerve terminal located a bit distant from muscle is preferentially affected at first in GBS. Moreover, given that not all AMAN patients showed CK elevation, a specific mechanism other than those directly caused by axonal damage may also be necessary for CK elevation. Mentioned above, we found that AMAN following URTI, including with anti-GM1 antibody, may be a specific feature of GBS patients with elevated CK levels. Although we could not investigate the antecedent pathogens, AMAN following URTI, including with anti-GM1 antibody, was reportedly associated with a specific pathogen such as Haemophilus influenzae. Therefore, the specific pathogen not identified in this study may elicit not only URTI and subsequent AMAN but also additional mechanisms that are specific to elevation of CK levels.
Our study has several limitations . First, it is small and retrospective and includes only patients from Japan, where AMAN patients are more frequent than other Western countries. Further large prospective studies in various population groups are needed. Second, although serial electrophysiological examinations were performed, the number of nerves, timing, frequency, and period varied for each patient. In some patients, serial examinations were performed only a few times over a short period or not performed at all. Of nerve conduction examinations on any motor nerves, that on the right median nerve was most often performed in our study. Therefore, we only showed the results of nerve conduction examinations on the right median nerve. Further studies with more frequent electrophysiological assessment for more and predetermined nerves at predetermined time points over a longer time period could identify these features in greater detail and with greater confirmation. Third, as mentioned above, while pathogens of antecedent infections are associated with various clinical phenotypes in GBS, we did not perform the comprehensive test to identify pathogens of antecedent infection, such as serum antibody to C. jejuni. Fourth, the test for anti-ganglioside antibodies was performed; however, only antibodies against GM1 and GQ1b were assessed in a limited number of patients. Information about antibodies to a ganglioside complex consisting of GM1 and GalNAc-GD1a is especially important as the antibodies have been reported to be associated with a pure motor variant with RCF following respiratory infection.[36-38] Information about IgG subclass (IgG1 or IgG3) of anti-GM1 antibodies is also important because the IgG subclasses of AMAN-associated antibodies have been reported to be associated with antecedent infection and the speed of recovery of GBS. However, these were not investigated in this study.