Pregnant Women
Bryan S. Michalowicz, Anthony J. DiAngelis, M.
John Novak, William Buchanan, Panos N.
Papapanou, Dennis A. Mitchell, Alice E. Curran,
Virginia R. Lupo, James E. Ferguson, James
Bofill, Stephen Matseoane, Amos S. Deinard, Jr.
and Tyson B. Rogers
J Am Dent Assoc 2008;139;685-695
© 2010 American Dental Association. The sponsor and its products are not endorsed by the ADA.
C O V E R S T O R Y
N
variety of populations sug-
CON
IO
gest that only about one-quarter to sider it safe to provide dental care for pregnant
T
T
A
women, supporting clinical trial evidence is lacking. N
I
one-half of women receive any U
IN U
C
pregnancy. Only 26 percent of women in this pregnant women. We used data from a multi-
study had been advised by a health care profes- center randomized controlled clinical trial (the
sional to see a dentist. Collectively, these findings Obstetrics and Periodontal Therapy [OPT] Trial)
suggest that use of dental services during preg- conducted to determine if periodontal therapy in
nancy may be driven more by patients’ and den- pregnant women reduces the risk of preterm
tists’ attitudes than by economics or convenience. delivery.13 We reported that scaling and root
Attitudes and behaviors among dentists may planing was not associated with adverse medical
arise from fear of causing harm to the pregnant events.13 Here, we report associations between
woman or fetus, fear of litigation or patients’ safety adverse pregnancy outcomes and essential dental
concerns. Notably, however, Hilgers and treatment (EDT), anesthetic use during nonsur-
colleagues9 found only one case in which a dentist gical periodontal treatment and combinations of
was sued—unsuccessfully—for treating a pregnant EDT and periodontal treatment. To the best of
woman who subsequently delivered a stillborn our knowledge, this is the first report of safety
infant. Dental care providers also may defer treat- outcomes in a cohort of pregnant women under-
ment in pregnant women because of fear that the going periodontal therapy and EDT.
anesthetics, antibiotics and analgesics commonly
SUBJECTS AND METHODS
one to four visits, and they administered topical brief hospital visits (< 24 hours) for laborlike
or local (injected) anesthetics at their discretion. pains on the advice of the OPT Data and Safety
We monitored the 410 control-group women for Monitoring Board, which judged that false labor
safety during their pregnancy and treated them is both common in pregnant women and not a
with scaling and root planing after delivery. true adverse event.
The women received comprehensive perio- We included hospitalizations for any other
dontal examinations at baseline (13 weeks, 0 reason because the relevant institutional review
days’ to 16 weeks, six days’ gestation), at 21 to 24 boards considered them to be serious adverse
weeks’ gestation and at 29 to 32 weeks’ gestation. events, regardless of their relationship to the
All subjects also had monthly visits, during which study. Women were hospitalized during the trial
women in the treatment group received tooth pol- for a variety of reasons, including uncontrolled
ishings and oral hygiene instruction, and women diabetes, cholestasis, ovarian cysts, pre-
in the control group received brief examinations eclampsia, chorioamnionitis, pancreatitis and
only. pyelonephritis.
Dentists evaluated women in both groups for Physicians diagnosed fetal anomalies via pre-
EDT needs. We defined EDT needs as the pres- natal ultrasonography or at delivery. Two women
TABLE 1
Previous 185 (72.8) 150 (68.8) 136 (78.2) 140 (79.1) .07
Pregnancy
Mean (± SD §) 0.22 ± 0.51 0.25 ± 0.50 0.25 ± 0.68 0.38 ± 0.59 .02
Number of
Medication Classes
(0-4)
Mean (± SD) 41.2 ± 24.1 42.7 ± 24.6 42.8 ± 26.7 43.4 ± 27.1 .83
Percentage of Sites
With CAL ¶ ≥ 2 mm #
Mean (± SD) 68.4 ± 17.4 70.1 ± 17.2 70.9 ± 17.0 68.7 ± 17.3 .43
Percentage of Tooth
Sites Bleeding on
Probing
Mean (± SD) Age 25.4 ± 5.3 25.5 ± 5.6 26.4 ± 5.7 27.0 ± 5.6 < .01
(Years)
* For a more detailed summary of the sample’s baseline characteristics, see Michalowicz and colleagues.13
† Unless otherwise specified.
‡ χ2 test P value reported for tabulated variables; analysis of variance P value reported otherwise.
§ SD: Standard deviation.
¶ CAL: Clinical attachment loss.
# mm: Millimeters.
Dental treatment and adverse outcomes. women who experienced any serious adverse
Table 1 lists the distribution of baseline charac- events was higher among those who received EDT
teristics (used in the propensity scores) among (12.1 percent for women who received partial
groups, according to EDT and periodontal treat- treatment and 10.3 percent for those who
ment status. The distribution of characteristics received completed treatment) compared with
differed significantly between groups (P < .05) for those who did not need this treatment (7.6 per-
clinical site, race, ethnicity, number of medication cent). These differences, however, were not statis-
classes and age. tically significant (P = .25), and they were even
Table 2 shows the number and percentage of less significant in the propensity score–adjusted
women who experienced a serious adverse event analyses (P = .59, Table 2). Rates of spontaneous
or an adverse pregnancy outcome, according to abortions/stillbirths, deliveries before 37 weeks’
the need for and receipt of EDT. We report gestation and fetal/congenital anomalies also
results from both unadjusted and propensity were similar in women who received partial or
score–adjusted analyses. The proportion of complete EDT. The adjusted odds ratios (ORs) for
TABLE 2
all adverse outcomes were close to 1, indicating The distribution of serious adverse events, spon-
that EDT was not associated with any significant taneous abortions/stillbirths, preterm deliveries
increase in risk for these events. and fetal/congenital anomalies did not differ sig-
Table 3 lists the proportion of women who nificantly between these groups (P > .05). The
experienced serious adverse events and adverse number of events in some treatment subgroups,
pregnancy outcomes, according to periodontal however, was small.
treatment status and anesthetic use. We present Table 4 (page 692) shows the rates of all serious
unadjusted ORs comparing rates according to adverse outcomes and adverse pregnancy outcomes,
anesthetic use and designate periodontally according to receipt of complete EDT and perio-
untreated control subjects as the referent group. dontal treatment, either treatment alone or neither
TABLE 3
treatment. As in Table 2, we present unadjusted the study results showed no significant differ-
and propensity score–adjusted ORs. Again, the ences in rates of any adverse event between
distribution of adverse outcomes did not differ women who received complete EDT, who needed
significantly between these groups. None of the but did not receive complete EDT, and who did
individual ORs—comparing events in groups not require EDT (all P values > .2; Fisher exact
receiving EDT and/or periodontal treatment with tests for three-way comparisons).
events in the group that received neither—was Of the 20 fetal or congenital anomalies, most
significantly greater than 1. were judged by an obstetrician (V.R.L.) to have
We also compared event rates between those occurred at the time of conception (for example,
who required but did not receive complete EDT chromosomal defects) or early in the first
and those who received complete EDT separately trimester before women were eligible for the trial.
in women who received and who did not receive Only three anomalies (fetal ventriculomegaly,
periodontal treatment. Few spontaneous abor- unspecified neuromuscular disorder and fetal car-
tions/stillbirths and fetal/congenital anomalies diac dysrhythmia) (Table 2) were judged to have
occurred in some of these subgroups (0 to seven). possibly occurred after subjects were eligible for
We found no significant differences between EDT study enrollment. The rates of these latter events
groups for women who received or who did not did not differ significantly (P > .05) between
receive periodontal treatment at a time before 21 groups, according to EDT or periodontal
weeks’ gestation (all P values > .3; Fisher exact treatment.
tests). Similarly, for serious adverse events and
deliveries before 37 weeks’ gestation, there were DISCUSSION
no significant differences between EDT groups for In this population, periodontal treatment and
women who received or who did not receive perio- EDT, administered at 13 to 21 weeks’ gestation,
dontal treatment (Table 5, page 693). Moreover, did not significantly increase the risk of any
TABLE 4
adverse outcome evaluated. With regard to ciated with routine dental care during
preterm birth risk, our findings are consistent pregnancy.20 Experts recommend that pregnant
with those of Lydon-Rochelle and colleagues,4 who women defer elective care before eight weeks’ ges-
found that rates of preterm deliveries (< 37 tation, when major organogenesis occurs, as well
weeks’ gestation) were similar in women who as during late pregnancy to avoid supine hypoten-
reported having received or not having received sion and general discomfort.9,10 Many obstetri-
dental treatment during their pregnancy, regard- cians, however, believe that dentists are overly
less of whether the care was problem-directed or cautious about providing dental care to pregnant
preventive. women.20
Clinicians’ views. The consensus in the In a survey conducted in the early 1990s, most
obstetrics community is that few risks are asso- obstetricians responded that they would like to be
TABLE 5
EDT Not EDT Needed EDT EDT Not EDT Needed EDT
Needed But Not Completed Needed But Not Completed
Completed Completed
consulted before a dentist provides “some” routine Randomized controlled trials. In addition
dental treatment (79 percent), provides treat- to the OPT trial, four randomized controlled trials
ments that induce a bacteremia (79 percent) or of periodontal treatment in pregnant women have
prescribes an antibiotic (88 percent).20 Ninety-one been conducted.13,14,24,25 Jeffcoat and colleagues14
percent of respondents did not want to be con- did not report safety outcomes for 366 women
sulted before all routine dental care. The vast randomized into a trial comparing preterm birth
majority of respondents did not think that stress rates among women receiving a prophylaxis or
related to routine dental care is hazardous to the scaling and root planing, with and without sys-
mother (95 percent) or fetus (97 percent). Among temic metronidazole therapy. Lopez and col-
general dentists, substantial proportions did not leagues15 randomized 400 women to receive
consult an obstetrician before administering rou- scaling and root planing at a time before 28
tine (78 percent) or emergency (50 percent) weeks’ gestation or after delivery. Eight women in
dental care.21 the treatment group and six women in the control
Dentists and patients may elect to defer care group experienced spontaneous abortions in this
during pregnancy when, in fact, pregnant women study, suggesting that periodontal intervention
may require more, not less, dental care.9 Changes did not substantially increase the risk of experi-
in local (tissue) and systemic estrogen levels encing a spontaneous abortion.
during pregnancy cause vascular changes and Sadatmansouri and colleagues24 conducted a
qualitative changes in the subgingival oral micro- trial that included only 30 pregnant women, and
biota that can lead to increased gingival bleeding they did not report safety data. Most recently,
and exuberant soft-tissue reactions to local irri- Offenbacher and colleagues25 reported two “fetal
tants.22,23 Nausea and vomiting may predispose a demises” among 67 randomized women for whom
pregnant woman to experience periods of dental birth outcome data were available. Neither the
erosion. Thus, guidelines12,16 generally advocate timing nor the group assignment (scaling and root
continued preventive and routine dental care planing at a time before 22 weeks’ gestation or six
throughout pregnancy. weeks postpartum) was specified for these events.
We know of no randomized controlled trials or by the potentially stronger risk factor of periodon-
prospective cohort studies of the relationship titis. Several lines of evidence, however, suggest
between nonperiodontal dental care, such as otherwise. We recruited from clinics that serve
EDT, and pregnancy outcomes. predominantly minority women who are, in gen-
We explored associations between dental treat- eral, at higher risk than are non-Hispanic whites
ment and adverse pregnancy outcomes in all study of experiencing adverse pregnancy outcomes.27 The
subjects, including those without EDT needs rate of stillbirths in black women in this study
(Tables 2 and 4) and in only those with EDT needs (10 [2.8 percent] of 358) was higher than the
(Table 5). Regardless of how we grouped women national rate for this group (1.2 percent).27 This
according to EDT needs and provision of care, was expected, however, because one of two sites
EDT was not associated with significantly higher that enrolled predominantly black women re-
risks of experiencing serious adverse events, spon- cruited from both high-risk and general obstetrics
taneous abortions or stillbirths, deliveries before clinics. Our stillbirth rates in non-Hispanic and
37 weeks’ gestation, or fetal or congenital anom- Hispanic whites were similar to national averages.
alies. Notably, our results also were qualitatively Also, the rates of adverse events, according to
identical when we used a propensity score that EDT, were similar for groups that received or did
risk of experiencing these adverse events and out- 5. Al Habashneh R, Guthmiller JM, Levy S, et al. Factors related to
utilization of dental services during pregnancy. J Clin Periodontol
comes. Additional large retrospective and 2005;32(7):815-821.
prospective studies, as well as studies of other 6. New York State Department of Health. Oral health plan for New
York State. Albany, N.Y.: New York State Department of Health;
dental treatments, are needed to confirm the August 2005. “www.health.state.ny.us/prevention/dental/docs/
safety of dental care during pregnancy. oral_health_plan.pdf”. Accessed May 6, 2008.
7. Pistorius J, Kraft J, Willershausen B. Dental treatment concepts
for pregnant patients: results of a survey. Eur J Med Res 2003;8(6):
Dr. Michalowicz is the Erwin Schaffer Chair in Periodontal Research 241-246.
and a professor of periodontics, University of Minnesota School of 8. Lindow SW, Nixon C, Hill N, Pullan AM. The incidence of maternal
Dentistry, 17-116 Moos Tower, 515 Delaware St. S.E., Minneapolis, dental treatment during pregnancy. J Obstet Gynaecol 1999;19(2):
Minn. 55455, e-mail “micha002@umn.edu. Address reprint requests 130-131.
to Dr. Michalowicz. 9. Hilgers KK, Douglass J, Mathieu GP. Adolescent pregnancy: a
Dr. DiAngelis is chief of dentistry, Department of Dentistry, review of dental treatment guidelines. Pediatr Dent 2003;25(5):
Hennepin County Medical Center, Minneapolis. 459-467.
Dr. Novak is a professor of periodontics and associate director of the 10. Little JW, Falace DA, Miller CS, Rhodus NL. Dental Management
Center for Oral Health Research, College of Dentistry, University of of the Medically Compromised Patient. 7th ed. St. Louis: Mosby;
Kentucky, Lexington. 2008:270, 273-274.
Dr. Buchanan is an associate professor of periodontics, Department 11. Bearfield C, Davenport ES, Sivapathasundaram V, Allaker RP.
of Periodontics and Preventive Sciences, University of Mississippi Possible association between amniotic fluid micro-organism infection
Medical Center, Jackson. and microflora in the mouth. BJOG 2002;109(5):527-533.
Dr. Papapanou is a professor of dental medicine, Division of Perio- 12. Edwards C, Yi CH, Currie JL. Chorioamnionitis caused by
dontics, Section of Oral and Diagnostic Sciences, Columbia University Capnocytophaga: case report. Am J Obstet Gynecol 1995;173(1):