Page 146 - JSOM Winter 2018

FIGURE 3 Funnel plot of studies examining the influence of FIGURE 5 Funnel plot of studies examining the influence of
glucosamine sulfate on pain. glucosamine sulfate on pain.

Std diff, standardized difference. Std diff, standardized difference.
FIGURE 4 Forest plot of studies examining the influence of
glucosamine sulfate on joint space width. schedule, study length, and methodological quality of the
studies. The stratified analysis showed that studies with in-
dustry involvement reported a greater reduction in pain than
studies in which industry was not involved. This may suggest
bias, but studies without industry involvement still showed
that GS reduced pain, albeit with an effect size less than half
that of the industry-involved investigations. Studies without
industry involvement had reduced (although still significant)
heterogeneity, suggesting the actual SMD may lie closer to the
SMD associated with those studies. Besides bias, the possibil-
ity that some methods for preparation of GS may be more
effective than others cannot be ruled out. All the studies with
GS, glucosamine sulfate; Jt Sp, joint space; Std diff, standardized industry involvement involved the use of crystalline GS. One
difference. study without industry involvement also used crystalline GS
and showed significant pain reduction efficacy. Inclusion of
74
(Table 2) so a fixed model was used. There was little vindica- that study with the industry involved investigations (n = 7
74
tion of publication bias (Table 2). Stratified analysis (Table 2) studies, summary SMD = –0.57, 95% CI = –0.88 to –0.27)
indicated larger effects for studies involving industry and and exclusion from the non–industry-involved studies (n = 10
much less for those not industry involved. studies, summary SMD = –0.21, 95% CI = –0.40 to –0.02)
had little effect on the summary SMDs. Crystalline GS is pro-
Discussion duced by a process patented in the United States by the Rotta-
79
pharm Group (Monza, Italy). It is available under the brand
The present analysis indicated that orally consumed GS had names Dona , Viatril-S , Arthryl , Osaflexan , Xicil , and
®
®
®
®
®
a small to moderate effect on reducing OA-related pain and Glusartel , although in the United States, only the first two
®
little effect on joint-space narrowing. The pain findings are listed products appear to be available. In the past, there were
relatively consistent with those of other reviews 34,36,42 that sep- problems with the purity of DSs claiming to contain GS, but
80
arately analyzed studies using GS versus other glucosamine purity has improved substantially. 81,82 In 2002, eight examined
formulations, although the other reviews included studies in- products contained only 63% (20%) of the labeled glucos-
volving routes of administration other than oral. Similar to amine content; in 2014, these same products contained 94%
other reviews, 34,36 the present review found a very large degree (13%). Future studies could examine the efficacy of crystal-
81
of heterogeneity among these pain studies, but this was to be line GS versus those manufactured in other ways.
expected, given the variations in study designs, dosages, length
of treatments, and types of pain measures. The few previous The finding that several GS administrations per day had larger
78
systematic reviews that examined joint-space changes with GS effects on pain reduction than a single large dose (1,500mg/d)
treatment were earlier ones 46,48 that included only the initial suggests that the dosing schedule is important. The mechanism
positive efficacy studies 64,65 and not the later studies showing of action whereby GS reduces pain may involve a reduction in
minimal joint structural efficacy. 60,66 Heterogeneity was very inflammation at the joint. When GS is orally administered, C
14
low in these joint-space studies, likely because of the similar- tracer studies have shown that 90% is absorbed and quickly
ities in the outcome measures (joint widths determined from enters numerous tissues, including the articular cartilage. In
83
radiographs) and longer length of treatment. Publication bias isolated human cartilage, crystalline GS reduced proinflamma-
was also minimal because of the larger (and similar) sample tory factors (namely, interleukin-1, tumor necrosis factor, cy-
sizes and relatively narrow range of SMDs. clooxygenase-2) and decreased the expression of other factors
that degrade the cartilage matrix (e.g., matrix metalloprotein-
In interpreting the pain results, several factors should be ases). 84–87 Anti-inflammatory effects in isolated cartilage can be
taken into account, including industry involvement, dosage achieved at concentrations of about 10μmoles/L, but it is not
84
144 | JSOM Volume 18, Edition 4 / Winter 2018