Histologie Evaluation of New Attachment Apparatus

Formation in Humans
Part I
Gerald M. Bowers,* Brian Chadroff,f Robert Carnevale,t
James Mellonig,§ Rüssel Corio, | Jane Emerson,* Mark Stevens,H
and Elaine Romberg#

Part i of this three-part human study evaluated the formation of a new attachment
apparatus (bone, cementum, and periodontal ligament) on pathologically exposed root
surfaces in an open and closed environment. The most apical level of calculus on the root
served as a histologie reference point to measure regeneration on root surfaces exposed to
the oral environment. Attempts were made to initiate the formation of a new attachment
apparatus by flap curettage, root planing, coronectomy, and submersion of vital roots
beneath the mucosa. Nonsubmerged defects were treated by the same surgical technique
and served as controls. Biopsies were obtained at 6 months and regeneration was evaluated
histometrically by two investigators who were unaware of the treatment performed. Data
from 9 patients with 25 submerged and 22 nonsubmerged defects were submitted for
statistical analysis. Results indicate that a new attachment apparatus did not form in any of
the 22 nonsubmerged teeth; a new attachment apparatus did form in a submerged environ-
ment (0.75mm); significantly more new attachment apparatus (P < 0.05), new cementum
( < 0.01 ), new connective tissue ( < 0.05), and new bone ( < 0.02) formed in submerged
defects; new cementum was cellular in nature and formed equally well on old cementum
and dentin. Greater percent positive regeneration of the attachment apparatus and all
component tissues occurred in submerged defects and no extensive root résorption, ankylosis,
or pulp death was observed on submerged or nonsubmerged roots. (Journal of Periodontol-
ogy, 1989;60:664-674)

The regeneration of a new attachment apparatus
(bone, cementum, and a functional periodontal liga-
ment) after the natural attachment has been destroyed
by disease or traumatic injury is one of the most chal-
lenging problems in dentistry. New attachment studies
have been conducted in both animals and humans.
Unfortunately, it is not always possible to extrapolate
the results of new attachment studies in animals to
humans. Evaluation of new attachment in artificially-
created defects in animals with a different bacterial
*
Department of Periodontics, Baltimore College of Dental Sur-
gery, University of Maryland, Baltimore, MD.
t Private practice, Rego Park, NY.
% Private practice, New London, CT.
§ Department of Periodontics, School of Dentistry, University of
Texas, San Antonio, TX; previously, Naval Dental Center, Periodon-
tics Department, Bethesda, MD.
I Department of Oral Pathology, Georgetown University, School
of Dentistry, Washington, DC.
H Department of Removable Prosthetics, Baltimore College of
Dental Surgery, University of Maryland, Baltimore, MD.
#Department of Educational and Instructional Resources.
664
flora than humans may produce misleading results.
Even when the optimal primate model system reported
by Caton and Kowalski1 was used, completely divergent
findings were reported by Caton and coworkers2"4 and
Mellonig.5 Caton was unable to promote new attach-
ment in primates with or without grafting materials,
whereas Mellonig demonstrated new attachment in
grafted sites and in a significantly fewer number of
nongrafted defects. Animal studies will continue to
provide useful information, but proof of formation of
a new attachment apparatus must come from histologie
evidence in humans. While evidence of the formation
of a new attachment apparatus has been reported in
humans,6 some question existing information on the
basis of inadequate documentation of denuded root
surfaces,7"9 poor histologie technique,8 or inaccurate
reference points for measurement,7"9 and contend that
such regeneration is not feasible in areas previously
exposed by Periodontitis. Others have also questioned
the value of graft materials and whether or not a graft
contributes to new attachment.10 It is apparent that
several important questions need to be answered re-
Volume 60
Number 12 Histological Evaluation ofNew Periodontal Attachment, I
garding the formation of a new attachment apparatus.
In this three-part human study, regeneration of intra-
bony defects was compared in submerged and nonsub-
merged environments, with and without grafting ma-
terials. In each of the three parts, the most apical level
of calculus on the root served as the histologie reference
point to measure regeneration.7 In Part I, attempts were
made to regenerate intrabony defects by: 1 ) flap curet-
tage, crown removal and submersion of the vital root
beneath the mucosa and 2) flap curettage alone. In Part
II, regeneration was compared in submerged intrabony
defects treated with and without decalcified freeze-
dried bone allograft (DFDBA). Part HI evaluated the
potential for regeneration of intrabony defects on non-
submerged teeth with and without the use of DFDBA.
In a preliminary report of this project, it was sug-
gested that a new attachment apparatus may form after
flap curettage and root planing of intrabony defects if
the vital roots are submerged beneath the mucosa.11 In
nonsubmerged teeth, using the same surgical technique,
a long junctional epithelium formed over the patholog-
ically exposed root surfaces. This paper will present the
final results of Part I and attempt to answer the follow-
ing questions: following flap curettage in intrabony
defects and root planing of pathologically exposed root
surfaces
1. Is it possible to regenerate a new attachment ap-
paratus on nonsubmerged roots?
2. Will a new attachment apparatus regenerate on
submerged vital roots?
3. Are there significant differences in the amount of
attachment apparatus and component tissues that will
form on submerged and nonsubmerged roots?
4. Will the attachment apparatus and component
tissues form more frequently on submerged or nonsub-
merged roots?
5. Will extensive root résorption, ankylosis, or pulp
death occur on submerged or nonsubmerged roots fol-
lowing flap curettage and root planing?
MATERIALS AND METHODS
Patient Selection
Thirteen volunteers were solicited from patients who
had two or more maxillary or mandibular premolars,
cuspids, or incisors recommended for extraction by the
Oral Diagnosis Department. The basis for extraction
was advanced periodontal disease, excessive mobility,
malpositioned teeth, prosthetic considerations, esthet-
ics, and patient request." Teeth chosen for the study
demonstrated advanced bone loss, deep pockets, and
associated interproximal intrabony defects with radio-
graphically visible calculus on the root surface. All teeth
were vital and asymptomatic. Volunteers were system-
ically healthy and instructed both orally and in writing
as to the nature of the project to obtain an informed
665
consent. This project was approved by the Human
Volunteers Committee, University of Maryland.
Presurgical Therapy
Treatment plans were developed and patients "re-
ceived periodontal and restorative care as necessary to
restore their mouth to a state of health. Calculus was
not removed from the teeth included in the study.
Surgical Therapy
All surgical procedures and clinical measurements
were performed by the same investigator (GMB). On
teeth randomly selected for vital root submersion (co-
ronectomy), a mucoperiosteal flap was reflected on the
facial surface and a gingivoplasty was performed on the
lingual or palatal surface. Vertical incisions were made
on the facial surface to permit coronal positioning of
the flap. A small hole was placed in the crown of each
tooth and suture material was passed through the hole
as a safeguard against aspiration of the crown once it
was severed from the root. Crowns were amputated at
the highest level of the alveolar crest with a #701 cross-
cut fissure bur in a high speed handpiece with a sterile
water spray to prevent soft and hard tissue damage.
Sharp edges of the roots were rounded with a finishing
bur. Granulation tissue was removed from all defects
without instrumenting the root surface. Once the gran-
ulation tissue was removed, a 1/4 round bur was used
to notch the root at the base of the calculus in order to
delineate exposed from nonexposed root surfaces.7 A
fiber optic light source was used to illuminate the defect
and the adjacent root surface and magnifying loops
were used to aid in locating the base of calculus. Root
planing was accomplished from the calculus reference
notch coronally by hand instrumentation and finishing
burs. Special care was taken to preserve the position of
the notch during root planing. An ultrasonic instru-
ment* was used for final debridement of the defect but
no attempt was made to contact the root surface apical
to the calculus reference notch. The bony defects were
classified as to the number of walls, and the depth of
each defect was measured with a calibrated periodontal
probe from the alveolar crest to the calculus reference
notch (groove) and from the notch to the depth of the
defect (Fig. 1). In nonsubmerged teeth, the distance
from the cemento-enamel junction to the alveolar crest
was also measured. The bony walls of the defects were
penetrated in numerous places with a 1/4 round bur
(intramarrow penetration) to enhance rapid prolifera-
tion of blood vessels into the defect. In order to obtain
soft tissue closure over the submerged roots, the facial
flap was coronally positioned and sutured over the
gingivoplasty wound. Vertical incisions were also su-
tured. The wound edges around the nonsubmerged
teeth were tightly adapted with vertical mattress sutures.
*
Cavitron, Dentsply Corporation, York, PA.

666 Bowers, Chadroff, Carnevale, Mellonig, Corlo, Emerson, Stevens, Romberg
AC- Notch at alveolar crest
Groove- Notch at base of calculus
BD- Base of intrabony defect
CEJ- Cementoenamel junction
Figure 1. Clinical Measurements.
An adhesive material was placed over the surgical sites
of nonsubmerged teeth and the area was covered with
a periodontal dressing. If a temporary denture was
placed at the time of surgery, it was relieved over the
submerged roots. A resilient denture material was used
to reline the denture in 3 to 4 weeks. All patients were
given Tetracycline, 250mg, 4 times a day for 14 days
while plaque control was compromised. A mild anal-
gesic such as Ibuprofen, 400mg, was also prescribed.
All defects and root surfaces were treated identically
except the nonsubmerged roots were notched at the
highest level of the alveolar crest as well as at the base
of calculus.
Postoperative Maintenance
Sutures were removed in 7 days and the sites were
debrided with saline. Nonsubmerged teeth were lightly
debrided and polished. Patients were seen on a weekly
basis for the first month, every 2 weeks for the next 2
months, and monthly for the 3 succeeding months for
light debridement, 0.4% stannous fluoride application,
and reinforcement of plaque control when indicated.
Temporary appliances were evaluated at maintenance
visits and relined when indicated. Color transparencies
were taken throughout the course of treatment. Radio-
graphs were retaken at the end of the 6-month period.
Obtaining the Biopsy Specimens
At the end of 6
months, a second surgical procedure
was performed to remove the submerged roots and
nonsubmerged teeth en bloc." Most studies indicate
Stomahesive, E. R. Squibb & Sons, Princeton, NJ.
J. Periodontol.
December 1989
that cementogenesis has peaked at 3 months;12"14 how-
ever, 6 months was selected in order to compensate for
individual patient differences in wound healing. The
coronal portions of the submerged roots were exposed
to establish orientation, but interproximal tissue was
left undisturbed. Mucoperiosteal flaps were reflected
and the roots were sectioned with a #700 L cross-cut
fissure bur in a high speed handpiece with sterile water
spray. A small interproximal wedge of tissue and a
section of root approximately 5 mm wide, 7 mm long
and 5 mm thick was removed.11 Biopsies were notched
on the crown or root stump parallel to the desired plane
of sectioning to aid in proper alignment at the time of
embedding. Biopsies were then placed in buffered 10%
formalin. Remaining root fragments were removed and
a ceramic material* was placed into the biopsy and
extraction sites to maintain the original shape and in
some instances to augment the existing ridge. The facial
and palatal tissues were then repositioned and sutured
over the sockets. Patients were seen in 7 days to remove
the sutures. After the ridges had healed, the patients
were referred to the Prosthetic Department for the
fabrication of permanent prosthetic appliances.
Histologie Technique and Evaluation
Specimens were processed at the Armed Forces In-
stitute of Pathology and the University of Maryland
School of Dentistry. Biopsies were decalcified in EDTA,
embedded in paraffin, and serially sectioned at 7 µ .
Five submerged and 5 nonsubmerged specimens were
embedded in plastic. Every ninth section was stained
with Masson's trichrome or Goldner's stain and every
tenth section with hematoxylin and eosin. A blind
histologie evaluation of all biopsies was conducted by
an oral pathologist (RC) and a periodontist (JM), to
determine the measurements shown in Figures 2 and
3. It should be noted that the most coronal aspect of
submerged roots served as reference point A, since the
roots were amputated at the highest level of the alveolar
crest at the time of surgery. Measurements were made
simultaneously using a dual attachment binocular mi-
croscope, with a micrometer reticle.§ All serial sections
with readily identifiable reference points and suitable
hard and soft tissue relationships were measured by the
examiners. Magnification was altered to verify specific
tissue types and relationships but all measurements
were made at 35 X. Examiners agreed on each meas-
urement before it was recorded. If there was disagree-
ment between examiners, both measurements were in-
cluded and averaged. A mean score was determined for
each measurement shown in Figures 2 and 3. In sub-
*
Synthograft, Johnson & Johnson Corporation, New Brunswick,
NJ.
§
Micrometer Reticle, American Optical Corporation, Buffalo, NY.
Volume 60
Number 12 Histológica! Evaluation ofNew Periodontal A ttachment, I 667
merged sites, if there was evidence of epithelial "break-
through" (presence of a junctional epithelium), data
were collected but not included in this study. This was
necessary to insure that regeneration was evaluated in
a closed environment free of influence of junctional
epithelium and oral microflora. The examiners also
noted evidence of extensive root résorption, ankylosis,
pulp death, type and location of cementum formation
(over old cementum, dentin, or both) and orientation
of the periodontal ligament fibers (parallel, perpendic-
ular or both).
Number of Patients and Defects
During the 4 years of the project, two patients with-
drew from the study because of work requirements.
One submerged defect was not evaluated due to the
lack of a definitive reference notch. All submerged and
nonsubmerged defects were lost in one patient due to
Figure 2. Histological measurements on nonsubmerged defects technician error in sectioning. Three submerged defects
NAA New attachment apparatus formation demonstrated epithelial breakthrough and were ex-
NC New cementum -

cluded from the study. A total of 25 submerged and 22

New bone
-

AC Alveolar crest
-
nonsubmerged defects in 9 patients were submitted for
CT Width of connective tissue attachment
-
statistical analysis (Table 1).
JE Junctional epithelium
-

A Notch at clinical location of alveolar crest Statistical

Analysis
-

Notch at base of calculus

-

D Dentin
-

Repeated Measures General Linear Model Analyses

OC Old cementum
-

of Variance was used to test for significant differences

PL Periodontal ligament in amount of new attachment apparatus and compo-
-

BP Bone particle nent tissues that formed in experimental and control

-

sites. Serial sections of each defect were evaluated for

positive regeneration of a new attachment apparatus
8evered Root
A Table 1
Histological Measurements in mm: Comparison of Flap Curettage
in 25 Submerged and 22 Nonsubmerged Teeth, 9 Patients

Measurement
Submerged Nonsubmerged F
„

NAA 0.75±0.86f 0.00 ± 0.00 7.87 .05

NC 1.14 ±0.89 0.01 ±0.04 13.97 .01
CT 1.49 ±1.09 0.01 ±0.11 6.07 .05
NC and CT 2.51 ±0.80 0.02 ±0.14 78.57 .0001
NB 0.96 ± 0.90 0.02 ±0.12 10.00 .05
B- 2.58 ± 0.74 2.26 ± 0.64 1.16 .33
B-AC 0.99 ± 0.89 0.11 ±0.28 5.05 .07
B-JE 0.00 ±0.00
A-AC -1.46 ±0.93 -2.15 ± 1.08 3.21 .13
— —
—

AC-JE -0.30 ±1.39

— — —

Figure 3. Histological measurements on submerged defects NAA New attachment apparatus.

=

NAA New attachment apparatus formation NC New cementum.

=

NC New cementum -

CT Width of new connective tissue.

=

NB New bone
-

NC and CT New cementum plus new connective tissue attach-

=

AC Alveolar crest
-

ment.
CT Width of connective tissue attachment
-

NB =New bone.
A Root severed at highest level of alveolar crest
-
=
Reference notch at base of calculus.
Groove at base of calculus
-

A =
Reference notch at alveolar crest (nonsubmerged); severed
D Dentin
-

root surface at alveolar crest (submerged).

OC Old cementum
-

JE Most apical level of junctional epithelium.

=

PL Periodontal ligament
-

AC Microscopic location of alveolar crest,

=

BP Bone particle
-

t mean ± standard deviation (mm).

-
 J. Periodontol.
668 Bowers, Chadroff, Carnevale, Mellonig, Corio, Emerson, Stevens, Bömberg December 1989

and component tissue formation. The percentage of

positive regeneration for each technique was reported.
The most frequent location of new cementum forma-
tion, and the most common orientation of the peri-
odontal ligament fibers were also reported for each
technique.

RESULTS

Is It Possible to Regenerate a New Attachment Appa-

ratus on Nonsubmerged Roots?
The formation of a new attachment apparatus was
not observed in any of the 22 nonsubmerged teeth
treated by flap curettage and root planing (Table 1,
Figs. 4 and 5). On occasion, a small amount of new
cementum (xO.Ol mm), new connective tissue (xO.Ol
mm), or new bone formation (x0.02 mm) was observed
coronal to reference notch (Figs. 6 and 7). In no
instance, however, did all three tissues regenerate si-
multaneously to form a new attachment apparatus. In
nonsubmerged defects, the junctional epithelium was
usually located at or slightly below the alveolar crest
and the calculus reference notch (B-AC xO. 11 and =
Figure 5. Higher magnification of Figure 1 (region of arrow) dem-
AC-JE =
x-0.30). onstrating junctional epithelium/connective tissue/cementum inter-
face. (H&E, original magnification, 40x)

Figure 4. Representative section of nonsubmerged defect after flap

curettage and root planing. Note that location of the junctional epithe-
lium (JE) is apical to the calculus reference notch (B) and the alveolar
crest. (H&E, original magnification, 4x). Arrow denotes junctional
epithelium/cementum/connective tissue interface shown at higher
magnification in Figure 2.
Figure 6. Nonsubmerged defect demonstrating new cementum for-
mation (arrow) at base of calculus notch (B). (H&E, original magni-
fication, 4x). Higher magnification of region of arrow is shown in
Figure 4.
 Volume 60
Number 12 Histological Evaluation ofNew Periodontal Attachment, I
Will a New Attachment Apparatus Regenerate on Sub-
merged Vital Roots?
A new attachment apparatus did form on pathologi-
cally exposed root surfaces in a submerged environment
(x0.75mm) (Table 1, Figs. 8, 9, and 10). Complete
regeneration of the attachment apparatus, B- (x2.58
mm), was limited by the amount of bone and cemen-
tum formation (x0.96mm and 1.14mm respectively).
Figure 7. Higher magnification ofcalculus notch (B) and new cemen-
tum (NC) formation shown in Figure 3. New attachment apparatus
formation was not evidenced coronal to calculus reference notch.
(H&E, original magnification, 40x)
Figure 8. Submerged defect demonstrating the formation of new
attachment apparatus from calculus reference notch (B) to arrow. New
cementum formation was observed at approximately the same level
as new bone formation (Masson's trichrome, 4x). Higher magni-
fication of calculus notch and region of arrow are shown in Figures
6 and 7.
669
Pathologically exposed root surfaces were most often
covered by a combination of cementum and connective
tissue (x2.51 mm) (Fig. 10).
New cementum was usually cellular in nature, varied
in thickness from section to section, and was often
separated from the root surface. This separation was
considered artifact and related to acid décalcification
and paraffin embedding, since the separation was not
observed in the 10 specimens which were decalcified in
EDTA and embedded in plastic. In submerged defects,
new cementum most frequently formed over both old
cementum and dentin within the same defect (Table 2,
Fig. 9). Periodontal ligament fibers were embedded in
cementum and bone and were most frequently oriented
parallel to the root (Table 3, Fig. 11).
Are There Significant Differences in the Amount of
Attachment Apparatus and Component Tissues That
Will Form on Submerged and Nonsubmerged Roots?
Data in Table 1 reveal that
significantly more attach-
ment .05), cementum ( < .01), con-
apparatus ( <
nective tissue (P .05), cementum plus connective
<
tissue (P < .0001), and bone (P < .05) formed in
submerged defects than nonsubmerged defects. There
was a difference in the mean location of the alveolar
crest from the calculus reference notch in submerged
Figure 9. Higher magnification of calculus notch (B) in Figure 5
demonstrating the formation of new cellular cementum (NC), new
bone (NB) and new periodontal ligament (PL). Note parallel arrange-
ment of periodontal ligament fibers and incorporation of Sharpey's
fibers in bone and cementum. Also note that new cementum formed
over old cementum and dentin. Space between new cementum and
root is artifact. (H&E, original magnification, 40x)
 J. Periodontol.
670 Bowers, Chadroff, Carnevale, Mellonig, Corio, Emerson, Stevens, Bömberg December 1989

Will the Attachment Apparatus and Component Tissues

Figure 10. Higher magnification of Figure 5 at level of alveolar crest
(AC) demonstrating new cementum (NC)/connective tissue (CT) in-
terface (arrow). Space between cementum, connective tissue and root
surface is artifact. (H&E, original magnification, 40x)
Form More Frequently on Submerged or Nonsub-
merged Roots?
Positive regeneration of the attachment apparatus,
cementum, connective tissue, cementum plus connec-
tive tissue, and bone occurred more frequently on
submerged roots than on nonsubmerged roots (Table
4).
Will Extensive Root Resorption, Ankylosis, or Pulp
Death Occur on Submerged or Nonsubmerged Root
Following Flap Curettage and Root Planing?
No extensive root résorption, ankylosis, or pulp death
was observed on submerged or nonsubmerged roots.
Surface résorption of old cementum and dentin was a
common finding on submerged roots and appeared to
precede cementogenesis (Fig. 12).
There significant difference between mean
was no
histological distances of submerged and nonsubmerged
reference notches B- (Table 1). Likewise, there was
no significant difference in probing measurements be-
tween alveolar crest and base of calculus ( -B), base of
calculus to the base of defect (B-BD), and alveolar crest
to the base of defect (A-BD) between submerged and
nonsubmerged defects (Table 5).

Table 2
Percent of Sections Showing Regeneration of New Cementum Over
Three Types of Tissue
Percent Percent Percent
Treatment over old over
V//-'
over
cementum dentin both r -
-1 " ,'j-Nc
Submerged 19% 19% 62%
Nonsubmerged 50% 50%
:
No new cementum formed.

Table 3 M'
Percent of Sections Showing Regeneration of Periodontal Ligament
in Three Directions
Percent Percent Percent
Figure 11. New attachment apparatus formation coronal to calculus
Treatment parallel perpendicular parallel and reference notch in submerged defect with parallel and perpendicular
to tooth to tooth perpendicular arrangement ofperiodontal ligament fibers (PL). Note new cementum
Submerged 50% 25% 25% (NC), new bone formation (NB), and incorporation ofSharpey'sfibers.
Nonsubmerged* —
(H&E, original magnification, 80x)
*
No new attachment formed
Table 4
Percent of Sections Showing Regeneration for Each Procedure
versus nonsubmerged sites (B-AC 0.99 mm vs 0.11 =
Measurement Submerged Nonsubmerged
mm respectively) but the difference was not significant. 0%
New attachment apparatus 62%
There appeared to be greater loss in height of alveolar New cementum 77% 3%
crest between submerged and nonsubmerged defects New connective tissue 77% 2%
cementum and connective tissue 90% 3%
(A-AC -1.46 mm vs -2.15 mm respectively) but the
= New
New bone 71% 3%
difference was not significant.
Volume 60
Number 12 Histological Evaluation ofNew Periodontal Attachment, I 671

to the calculus reference notch. A combination of ce-

mentum and connective tissue will form more fre-
quently and in greater quantity than cementum or
connective tissue alone. Some new bone formation may
also occur. Complete regeneration of a new attachment
apparatus in a submerged environment, however, ap-
pears to be limited by the amount of new bone and
cementum formation. These findings concur with
Cooke et al.15 who noted that the defects associated
with vitally submerged roots did not completely fill
with bone. Likewise, cementum was observed lining
the root surface at the base of only 2 of the 6 successfully
submerged roots. It also supports the findings of Got-
tlow et al.16 who noted limited bone formation when
membranes were placed to exclude the epithelium.
Epithelial exclusion then is an important factor in the
formation of a new attachment apparatus, but does not
appear to be the only factor necessary to enhance
complete regeneration of periodontal tissues.
Bone Formation
Minimal bone formation will occur from the calculus
reference notch coronally in nonsubmerged defects.
Even if defects are submerged, bone formation will be
limited in amount (x0.96 mm) and frequency of for-
Figure 12. Submerged defect with surface résorption of root coronal mation (71%). Newly formed bone is distinguishable
to calculusreference notch. Note cementoid formation (NC) apical to from existing bony walls by the presence of reversal
résorption bays (R). Also note new bone formation (NB) adjacent to lines, staining characteristics, and close approximation
cementoid formation. (Í1&E, original magnification. 40x)
to pathologically exposed root surfaces (Figs. 8, 11, and
Table 5
12).
Mean Clinical Measurements (mm) at Time of Surgery Cementum Formation
Measurement
Submerged Nonsubmerged
mean mean New cementum may form over old cementum, den-
A-B 3.11 ±1.33+ 3.21 ±0.92 0.44 .53 tin, or both old cementum and dentin in a submerged
B-BD 2.84 ±0.91 2.65 ±1.28 1.06 .34 environment. In 1963, Morris17 suggested that exposed
A-BD 5.95 ± 2.03 5.86 ± 1.80 2.18 .18 dentin may interfere with new cementum formation.
A =
clinical height alveolar crest Wilderman and Wentz18 observed that new cementum
=
base of calculus; BD base of defect
= formation was greatest when dentin was exposed. The
t Mean ± standard deviation (mm) results of our study suggest that cementum may form
equally well over old cementum and dentin. It has also
DISCUSSION been suggested that cementum formation is a response
to notch placement.17 Others have demonstrated that
This study compared the regenerative potential of the presence of a notch per se did not result in increased
the attachment apparatus and component tissues in cementum formation in dogs.19 It appears from the
intrabony defects in submerged and nonsubmerged en- results of our findings that notch placement has no
vironments. The data obtained suggest several impor- direct effect on the location or amount of new cemen-
tant clinical applications. tum formation.
New cementum formation was usually cellular in
Attachment Apparatus Versus Junctional Epithelium nature with incorporation of Sharpey's fiber attachment
Epithelial exclusion is an important factor in the type (Figs. 9 and 11 ). Stahl suggested that acellular cemen-
of attachment that forms after flap curettage and root tum is more conducive to fiber attachment than cellular
planing procedures. In nonsubmerged defects, the for- cementum.20 The incorporation of Sharpey's fibers was
mation of a long junctional epithelium is to be expected a consistent finding of cellular cementum in our study.
after flap curettage and root planing. If the epithelium Hence, the type of cementum formation may not be a
is excluded, new cementum, new connective tissue, or limiting factor in fiber attachment as previously sug-
new cementum and connective tissue will form coronal gested.

672 Bowers, Chadroff, Carnevale, Mellonig, Corlo, Emerson, Stevens, Bömberg
Periodontal Ligament Formation
In a submerged environment, the periodontal liga-
ment may be oriented parallel, perpendicular, or par-
allel and perpendicular in the same defect. Sharpey's
fibers will be incorporated in bone and cementum in
both parallel and perpendicular arrangements (Fig. 11).
Melcher stated that functional (perpendicular) orienta-
tion is unlikely without occlusal forces acting on the
teeth.21 Occlusion was not a factor in submerged vital
roots. It is possible, however, that the forces created by
dentures may have been sufficient to stimulate the
functional orientation in some submerged defects.
Pulpal Reaction to Coronectomy and Root Submersion
Pulp death does not appear to be a common sequela
of coronectomy. It also does not appear to be a sequela
of debridement and root planing of deep intrabony
defects, which in some instances extended to the apex.
Pulp tissue was not always present for histological eval-
uation, but when it was observed, there was often
evidence of secondary dentin formation opposite the
calculus reference notch in both submerged and non-
submerged roots (Fig. 13). It is not clear how pulpal
tissue survives and functions after coronectomy. All
sections with pulpal tissue are presently being evaluated
by a pulp pathologist who may be able to provide some
explanation.
Root Resorption and Ankylosis
Based on the results of this study, extensive root
résorption and ankylosis are not common sequelae of
vital root submersion in humans. Several authors have
demonstrated extensive root résorption of reimplanted
teeth in animals.22"24 It has been demonstrated in ani-
mals that, if cells from the gingival connective tissue22
or the alveolar bone23 contact the root surfaces, exten-
sive root résorption will occur. Based on these studies,
we anticipated that root résorption and ankylosis would
be a common occurrence on submerged roots since
Figure 13. Dental pulp (P) of submerged defect demonstrating sec-
ondary dentin formation (SD). (H&E, original magnification, 40x)
J. Periodontol.
December 1989
gingival connective tissue was placed directly over cut
dentin surfaces and various combinations of bony walls
surrounded the submerged roots. It was encouraging
that in no instance was extensive root résorption of
dentin or cementum noted in this study. Surface ré-
sorption of both dentin and cementum, however, was
a common finding. Resorption bays were frequently
lined with cementoid or newly formed cellular cemen-
tum. Surface résorption was such a consistent finding
that it appeared to the examiners that it may be a
prerequisite for new cementum formation on previ-
ously exposed root surfaces.
It has been suggested from animal studies that the
cells responsible for root résorption originate from sur-
gical flaps.23 In humans, the placement of gingival
connective tissues in contact with dentin or cementum
does not appear to result in extensive root résorption.
In fact, new cementum formation was occasionally
observed over the severed dentin surface (Figs. 14, 15,
and 16). Likewise, cells originating in bone do not
appear to produce extensive root résorption. It is pos-
sible then, that cells derived from other sources, such
as bone, can express the phenotype for new cementum
formation. This hypothesis has been proposed by other
authors25-26 and warrants further evaluation in humans.
Figure 14. New bone formation did not occur coronal to calculus
reference notch (B) in this submerged defect. New cementum has
formedfrom calculus notch to cut surface ofdentin (D) and over dentin
(arrows). (H&E, original magnification, 40x) Higher magnification
of cementum forming over severed dentin surface is shown in Figures
12 and 13.
Volume 60
Number 12 Histological Evaluation ofNew Periodontal Attachment, I 673
Figure 15. Higher magnification of severed dentin surface in Figure
11. Note cellular cementum (NC) overlying dentin (D). (H&E, original
magnification, 40x)
.V '·; .
D odontal surgery. II. Connective tissue attachment levels after four
Figure 16. Higher magnification of severed dentin surface in Figure
11 demonstrating cellular cementum formation (NC) and connective
tissue attachment (CT) with no indication of extensive root résorption.
(H&E, original magnification, 40x)
CONCLUSIONS
1. In submerged intrabony defects treated by flap
curettage and root planing:
A. It is possible to form a new attachment appa-
ratus on previously exposed root surfaces.
B. Significantly more new attachment apparatus
and component tissues will form in submerged
defects than in nonsubmerged sites. The amount
of new attachment apparatus formed is limited by
the amount of new bone and cementum formation.
C. A new attachment apparatus and component
tissues formed more frequently, in submerged de-
fects than in nonsubmerged sites.
D. New cementum formed over old cementum
and dentin, but more often formed over old ce-
mentum and dentin in the same defect.
E. The periodontal ligament was oriented parallel,
perpendicular, or parallel and perpendicular in the
same defect.
F. Extensive root résorption, ankylosis, and pulp
death are not common sequelae of this treatmejit.
2. In nonsubmerged intrabony defects treated by flap
curettage and root planing:
A. The formation of a new attachment apparatus
was not observed on previously exposed root sur-
faces.
B. A small amount of new cementum, connective
tissue, or bone formed on previously exposed root
surfaces.
C. Root résorption, ankylosis, and pulp death are
not common sequelae of this treatment.
ACKNOWLEDGMENTS
This study was supported by Grant DE 06250 from the National
Institute of Dental Research; Research Grant M0095.001.003, US
Navy Tissue Bank, NMRI, NNMC; and research funds from the
Armed Forces Institute of Pathology.
REFERENCES
JG, Kowalski CJ. Primate model for testing periodontal
1. Catón
treatment features: II. Production of contralaterally similar lesions. /
Periodontal 1976;47:506.
2. Caton JG, Zander HA. Osseous repair of an inffabony pocket
without new attachment of connective tissue. J Clin Periodontol
1976;3:54, 1976.
3. Caton J, Nyman S. Histometric evaluation of periodontal sur-
gery. I. The modified Widman flap procedure. J Clin Periodontol
1980;7:212.
4. Caton J, Nyman S, Zander H. Histometric evaluation of peri-
regenerative procedures. J Clin Periodontol 1980;7:224.
5. Mellonig JT. Histologie evaluation of freeze-dried bone allo-
grafts in periodontal osseous defects. J Dent Res 1981. 60 (Spec Issue):
(Abstr 388).
6. Bowers GM, Schallhorn RH, Mellonig JT. Histologie evalua-
tion of new attachment in human intrabony defects, a literature
review. J Periodontol 1982;53:509.
7. Cole RT, Crigger M, Bogle G, et al. Connective tissue regener-
ation to periodontally diseased teeth. / Periodont Res 1980; 15:1.
8. Moskow BS, Karsh F, Stein SD. Histological assessment of
autogenous bone graft—a case report and critical evaluation. J Per-
iodontol 1979;50:291.
9. Gara GG, Adams DF. Implant therapy in human intrabony
pockets: A review of the literature. J West Soc Periodont Periodont
Abstr 1981;29:2.
10. Egelberg J. Regeneration and repair of periodontal tissues. /
Periodont Res 1987;22:233.
11. Bowers GM, Granet M, Stevens M, et al. Histologie evaluation
of new attachment in humans. A preliminary report. J Periodontol
1985;56:381.
12. FroumSJ, Thaler R, Scopp IW, Stahl SS: Osseous autografts.
II. Histologie response to osseous coagulum-bone blend grafts. J
Periodontol 1975;46:656.
13. Dragoo MR, Sullivan HC. A clinical and histological evalua-
tion of autogenous iliac bone grafts in humans. Part II. External root
résorption. J Periodontol 1973;44:614.
14. Dragoo MR. Regeneration of the Periodontal Attachment in
Humans. Philadelphia: Lea & Febiger, 1981.
15. Cooke RT, Hutchins LH Jr., Burkes EJ Jr. Periodontal osseous
defects associated with vitally submerged roots. J Periodontol
1977;48:249.
16. Gottlow J, Nyman S, Karring T, Lindhe J. New attachment
formation as the result of controlled tissue regeneration. J Clin
Periodontol 1984;! 1:494-503.

674 Bowers, Chadroff, Carnevale, Mellonig, Corlo, Emerson, Stevens, Bömberg
17. Morris ML. Healing of human periodontal tissues following
surgical detachment: The effect of two horizontal nicks in the depo-
sition of new cementum. Periodontics 1963; 1:147.
18. Wilderman MM, Wentz FM. Repair of a dentogingival defect
with a pedicle flap. J Periodontol 1965;36:218.
19. Garrett S, Bogle G, Adams D, Egelberg J. The effect of
notching into dentin on new cementum formation during periodontal
wound healing. J Periodimi Res 1981:16:358.
20. Stahl SS. Repair potential of the soft tissue-root interface. J
Periodontol 1977;48:545.
21. Melcher AH. Repair of wounds in the periodontium of the
rat. Influence of periodontal ligament in Osteogenesis. Arch Oral Bio!
1970;15:1191.
22. Nyman S, Karring T, Lindhe J, Planteu S. Healing following
implantation of Periodontitis affected roots into gingival connective
tissue. / Clin Periodontol 1980;7:394-401.
23. Karring T, Nyman S, Lindhe J. Healing following implanta-
tion of Periodontitis affected roots into bone tissue. J Clin Periodontol
J. Periodontol.
December 1989
1980;7:96.
24. Karring T, Nyman S, Lindhe J, Surial M. Potentials for root
résorption during periodontal wound healing. J Clin Periodontol
1984;11:41.
25. Melcher AH, Cheong T, Cox J, Nemeth E, Shiga A. Synthesis
of cementum-like tissue in vitro by cells cultured from bone: A light
and electron microscope study. J Periodont Res 1986;21:592.
26. Aukhil I, Simpson DM, Johnson MC, Koch G. Progenitor
cell kinetics during guided tissue regeneration in experimental peri-
odontal wounds. J Periodont Res 1988;23:107-117.
Send reprint requests to: Dr. Gerald M. Bowers, Department of
Periodontics, University of Maryland, Baltimore College of Dental
Surgery, 666 West Baltimore Street, Baltimore, Maryland 21201.
Accepted for publication June 14, 1989.