Johnson, KE et a, J Clin Periodontol 1998; 25: 316-321
The abstract’s conclusions are not supported by the data. The study was
designed as a 3-month crossover (control and experimental groups switching
at 3 months) to eliminate criticism that the study design did not rule out
randomly higher plaque and calculus formation between the two groups, a real
possibility given the small sample size. However, despite repeated patient
instruction, the differences between control and experimental groups were
not statistically significant (i.e. no differences), so the authors just ignored
it as a presumed anomaly. They also discarded all data relevant to lingual
calculus. This is highly suspicious since the data tables show that both
the lingual PI and interproximal GI were statistically and significantly higher
than facial and non-proximal surfaces. Despite having dismissed the undesirable
data, the study was unable to find any significant differences in the overall
calculus index between experimental and control groups.
Other Problems:
- Statistical power of the study is weak. N=29 (only 16 control &
13 experimental subjects).
USC biostatisticians calculate that 200 patients would be required to achieve
adequate statistical power to make confident conclusions regarding changes
in Gingival(GI) and Periodontal (PI) Indices.
- The rationale for Hydro Floss efficacy (inhibiting electrovalent bonding
of plaque to the tooth) is based on dated concepts with supporting citations
at least 24-38 years old. We now know that bacteria attach through a variety
of mechanisms, including electrovalent but also ionic, hydrostatic and Van
Der Waals forces. The authors further mislead readers by implying that the
Hydro Floss uses the same the hydromagnetic technique used to reduce lime
and scale in pipes. It doesn’t. The ionization of water in treated pipes is
continuous, not episodic as in oral irrigation. Further, crevicular fluid
are replaced every 5-15 minutes, rapidly normalizing the sulcular environment
post irrigation.
- Contradiction: The Gingival Index
(GI) and the Standard Loe & Silness Index data for facial, lingual,
mesial and distal sites is listed as highly significant (p=0.0091), but 3
month data for the same scores is reported as a 27.5% reduction and not
significant. - Contradiction: Plaque Index
(PI) Standard Turesky index: (F, L) reported as weakly significant
(p=0.0343), but 3 month data for same scores reported as 2.2% and not significant.
- Accretion Data: Study uses a new, n
on-standard index. (Watt index: F, MF, ML, L, DL)
A modified Watt index was re-measured after prophy, but comparative data
is absent.
The study reports a strong positive association between the Watt accretion
(calculus) index and the Turesky plaque index, but the Watt accretion index
is actually a combined calculus and plaque index. Since plaque is incorporated
into the Watt index, it’s hardly surprising to see it correlate with Turesky.
The Watt Index data are non-significant (p=0.593) To claim, as
the authors do that it "Approaches significance" is not the same as
being significant.
Oddly, and unlike the GI and PI indices, only the facial data were considered
for measuring accretion. This is highly suspicious, since despite instructing
patients to specifically irrigate the mandibular anterior teeth, the study
concluded: "Regardless of treatment or period, lingual surfaces had statistically
significantly higher plaque indices than facial surfaces. Difference in plaque
index was affected more by period than treatment." "… the overall reduction
in the indices occurring over the 2 trial periods was not necessarily due
to the treatment (magnetized) but possibly to the fact that water oral irrigation
was used or a placebo effect."
- The study claims a 64% reduction in calculus (3-month), but the underlying
calculations are based on statistically non-significant data. Such claims
are meaningless and deceptive.
Subscribe to the OraMedia Newsletter for Dental Self
Sufficiency: