Clin. Otolaryngol.

2001, 26, 134±142

Pleomorphic adenomas of the major salivary glands:

a study of the capsular form in relation to surgical management
A . J . W E B B  & J . W. E V E S O N y

Department of Surgery, University of Bristol, Bristol Royal In®rmary and yDepartment of Oral and Dental Science,
University of Bristol, Dental Hospital and School, Bristol, UK

Accepted for publication 1 November 2000

W E B B A . J . & E V E S O N J . W.
(2001) Clin. Otolaryngol. 26, 134±142
Pleomorphic adenomas of the major salivary glands: a study of the capsular form in relation to
surgical management
This was a retrospective study of 126 primary pleomorphic adenomas to correlate capsular characteristics with
tumour histopathology in relation to current surgical debate (parotidectomy versus local excision). Capsular
thickness was measured by micrometry and tumours classi®ed into subtypes (1±4). Evidence of ®ne needle
aspiration damage (needle tracks, infarction) was sought. Minimal changes were seen in eight tumours. Tumour
growth features (bosselations, enveloping) were present in 57% and 33%, respectively, also microinvasion (42%) and
tumour `buds' (12%). Parotid lesions possessed thicker capsules than submandibular tumours. There was little
correlation between capsular thickness and cellular structure. The signi®cant exception was large (> 25 mm)
hypocellular parotid tumours which had thinner capsules and could be vulnerable to operative rupture. In 110
standard operations (parotidectomy, submandibular gland excision), capsular exposure was evident in 81%. Field
irrigation is recommended to lessen the risk of tumour seeding. This study reaf®rms many elements of capsular
weakness and suggests that parotidectomy is the operation of choice.
Keywords pleomorphic adenoma capsule recurrence parotidectomy needle biopsy

The surgical management of pleomorphic adenoma has
evolved since the 1920s,1±3 but during the past 50 years, there
has emerged a continuing debate concerning both diagnosis4,5
and surgical management.6±9 Despite recent optimism for a
complete solution of problems surrounding this unusual
tumour,10 confusion and uncertainty remains.
There is a consensus that, in the major salivary glands,
pleomorphic adenomas are enclosed by a layer of ®brous
tissue often termed a `capsule',11±16 but there is disagreement
over the form, extent and thickness of this layer.17 It has long
been recognised that a capsule might be extremely thin and
that tumour satellites or `buds' may extend through it (Figs 1
and 2). Masson18 classi®ed the assumed malignant potential of
the pleomorphic adenomas on the basis of capsular form and
his ideas were supported.2,11,12,19±22
Subsequently the uncertainty of capsular integrity was
convincingly demonstrated by the exhaustive microscopical
Correspondence: Mr A. John Webb, Senior Research Fellow,
Department of Surgery, University of Bristol, Level 7, Bristol Royal
Infirmary, Bristol BS2 8HW, UK.
studies of Patey and Thackray,23 who con®rmed the existence
and connections of pleomorphic adenoma satellites. They
®rmly discounted tumour multicentricity. At that time, surgi-
cal pioneers had already rejected the hitherto popular opera-
tion of enucleation or local excision in favour of super®cial,
facial nerve conserving parotidectomy,6,21,24,25 but this impor-
tant research persuaded many other surgeons away from local
excisions.26,27 The reputed advantage of parotidectomy was
that it ensured an adequate margin around the tumour. Never-
theless, enucleation-local excision retained adherents.28±32
Some surgeons de®ned a procedure termed `extracapsular
dissection' for a clinically benign parotid tumour.33 This
`re®ned' operation was distinguished from simple enuclea-
tion. The impression given was that `popular enucleation' was
in the style of Hybbinette34 who favoured direct incision of the
exposed tumour, curettage of the contents and sharp dissection
of the visible capsule. This was probably a false assertion and
many reported series of enucleation from reputable centres
were closer to extracapsular dissection.35,36 To examine this
genuine surgical controversy a precise knowledge of capsular
morphology is essential. Variations in capsular thickness,

# 2001 Blackwell Science Ltd 134

 Pleomorphic adnenomas of the major salivary glands 135
Figure 1. Satellite nodule (bud): tumour capsule thickness (capsular
thickness) 150±110 mm. Haemotoxylin and eosin, 10.
Figure 2. Absent and very thin capsule: capsular thickness 7±
30 mm. Haemotoxylin and eosin, 40.
microinvasion and tumour buds have been described in qua-
litative terms but quantitative data is very limited.11,12,22,23,37
In addition, recurrence following surgery is a serious com-
plication and although, in experienced hands, the incidence is
low (1±2%), the mechanism is incompletely understood.
Also, it is alleged that ®ne needle aspiration cytology
(FNA), itself a controversial mode of diagnosis, may induce
haemorrhage, infarction and perversion of histological detail.
This caveat is relevant to our centre where FNA is widely
used.38,39 These controversies concern current surgical
debate, hence an in depth investigation was undertaken to
address two major areas of importance, namely:
(1) A study of capsular form, incorporating micrometric
measurements to supplement standard histopathology
of the tumour edge, i.e. capsular absence, microinvasion,
tumour buds, capsular lamellation and bosselation, in
respect of capsular vulnerability at operation.
# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142
Table 1. Distribution of tumours: UBHT 1984±1995
Total patients (tumours) 126
Total operative procedures 127
Pleomorphic adenoma 126
Primary parotid 106
Primary submandiublar 19
Recurrent parotid 4
Accessory parotid 1
(2) Seeking evidence of previous FNA and identifying any
microscopical disruption or distortion related to need-
ling.40±42
Materials, patients and methods
ac c u m u l at e d s e r i e s
Examples of pleomorphic adenoma affecting the major sali-
vary glands were identi®ed from the volume and computer
histology ®les of the United Bristol Health Care Trust (UBHT)
for the interval 1984±1995 and the microscopical slides with-
drawn.
The accumulated series is shown in Table 1. The number of
histological slides per tumour varied from two to eight
(average four); staining was by haematoxylin and eosin. It
is accepted that there are limitations within a study of this type
which examines sections taken in limited planes. A minority
of the specimens were marked with India Ink so it is feasible
that minor capsular artefacts arose during ®xation and cutting
so the degree of invasion or capsular damage might have been
under estimated. Nevertheless, our material is representative
of routine pathological practice and is commensurate with
current NHS resources. Particular attention was directed
towards the following:
1. Capsular characteristics: the presence or absence of the
capsule (Fig. 2); gross surgical trauma; capsular lamella-
tion and a pericapsular areolar space (Figs 3 and 4).
2. Possible tumour growth features: bosselation at the sur-
face; envelopment of capsular structures; epithelial mar-
gination; microinvasion of the capsule; satellite buds
(Figs 1,2,5±7). Tumour size (mm) was measured directly.
3. Evidence of recent and past distortion from ®ne needle
puncture, especially the presence of tumour infarction.
(Fig. 8).
measurement
The capsule was measured using a standard micrometer and
graticule (2 mm 1 Interval ± 0.01 mm). Calibrations were
made for microscope objectives  2.5,  6.3  40
(magni®cations  25,  63 and 400, respectively). The
favoured objective was  6.3 whence 10 graticule units repre-
sented 150 mm. At this power it was possible to measure from
136 A.J. Webb & J.W. Eveson

Figure 3. Lamellated capsule measuring 100 mm in thickness. Figure 6. Seifert type 2 (hypocellular) adenoma; epithelial mar-
Haemotoxylin and eosin,  40. gination with thin capsular thickness, 15±45 mm. Haemotoxylin and
eosin, 25.

Figure 4. Chronic inflammatory infiltration of capsule: capsular

thickness 150 mm±pericapsular `areolar space' 150 mm wide. Figure 7. Microinvasion of capsule by tumour: capsular thickness
Haemotoxylin and eosin,  16. 450 mm. Haemotoxylin and eosin, 25.

Figure 5. Enveloping between two bosselae: compressed ducts

within capsule: capsular thickness 30±75 mm. Haemotoxylin and Figure 8. Needle track within an `infarcted' pleomorphic adenoma
eosin, 16. measuring 600  700 mm. Haemotoxylin and eosin, 6.

# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142

 Pleomorphic adnenomas of the major salivary glands 137
7 mm (0.5 units) to 2000 mm (2 mm). For each tumour 10±20
measurements of capsular thickness were made and judged
to be representative of capsular variations. The readings
were reproducible. In three cases of pleomorphic adenoma
`enucleation', the capsule was absent. Statistical assessment
of mean, SD, SEM AND P-values was undertaken by the Stu-
dent's two tailed t-test using a Microsoft Excel computer
program.
gr a di ng
Histological grading of subtypes was attempted in 130
tumours, according to the style of Seifert et al.43
Results
b o s s e l at i o n
Bosselation is de®ned as a smooth bulging prominence at the
tumour margin. Adjacent bosselations tend to infold some of
the surrounding capsule and compress salivary tissue into the
outer region of the tumour: for this we have used the term
`enveloping' (Fig. 5). To evaluate bosselation and enveloping,
143 slides from 126 tumours were examined. In 76/126
(60.3%) bosselation was present and in 27/76 (35%) this
moulding was extreme. For capsular enveloping, 44/126
(34.9%) demonstrated the feature, while in 9/44 (20.4%)
the ®brous intrusion was prominent. Epithelial margination,
characteristically arranged at the outer limit of myxoid bos-
selae, was present in 62/126 (49.2%) and marked in 6/62
(9.6%) (Fig. 6).
m i c r o i n va s i o n a n d t u m o u r b u d s
Microinvasion of the capsule was visible in 53/126 (42%)
primary adenomas (Fig. 7); tumour buds were detected in
15/126 (11.9%) (Fig. 1). These buds were bounded by a thin
®brous capsule. The presence of an `areolar space' around the
neoplasm was sought in 143 slides. An incomplete space was
detected in 42/143 (29.3%) but in only 3/42 (7.1%) was this
feature prominent and in no case was a total circumferential
areolar space seen (Figs 4 and 6).
m i c r o s c o p i c a l f e at u r e s
Microscopical features which might be attributed to direct,
trans®xed, FNA were problematic. Within 126 primary pleo-
morphic adenomas there were 10 possible intrinsic needle
tracks (8.0%) (Fig. 8). In 4/126 pleomorphic adenomas (3.1%)
minor tumour infarction was seen. Associated with 1/10
(10%) of possible needle tracks there was infarction. Assum-
ing that in 14/126 (11%) primary tumours there existed the
possibility of FNA-induced damage, in none was the histo-
# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142
Table 2. Tumour size related to capsular thickness measurements
Tumours Size (mm)

Small (> 8 mm/< 25 mm diameter )
Mean 223.8
SD 235.0
SEM 15.0
Large (> 25 mm diametery)
Mean 2020
SD 252.6
SEM 15.1

No. ˆ 20. Total capsular measurements 245; mean 12/tumour.
yNo. ˆ 20. Total capsular measurements 315; mean 15/tumour.
P ˆ 0.30 (not significant).
pathological characteristics of the lesion altered. FNA was
performed in 120/126 (95%) of tumours.
t u m o u r c h a r ac t e r i s t i c s
Several aspects of capsular thickness and tumour form were
examined and are presented in order. The range of capsular
thickness was 15±1750 mm. A comparison was made between
101 parotid and 19 submandibular pleomorphic adenomas.
Maximal, minimal and mean readings of capsular thickness
were submitted to statistical analysis. A signi®cant difference
emerged between those two sites. Parotid tumours possessed a
thicker capsule (P < 0.001).
Tumour size was assessed by direct measurements from the
histological sections and compared with capsular thickness.
Where multiple slides were available for one tumour, the
maximal size was selected. For each tumour the total number
of measurements was compared. It was appropriate to select
20 of the smallest and 20 of the largest tumours and the results
are presented in Table 2. We were unable to con®rm any
relationship between tumour size and capsular thickness.
The largest adenoma measured 35  25 mm; the smallest
8  8 mm.
Focal capsular absence and tumour cellularity was studied
by comparing hypercellular (Seifert subtypes 3 and 4) (Fig. 9)
with hypocellular (Seifert subtype 2) adenomas (Fig. 10),
yielding the following results: the hypercellular tumours (n
ˆ 38) possessed focal absence in 34%, whereas for hypocel-
lular tumours (n ˆ 5) the equivalent was 74%. There was a
clear association between cellularity and a visibly intact
capsule.
The relationship between tumour cellularity and capsular
thickness was examined in submandibular and parotid ade-
nomas. Maximal and total thickness measurements for Seifert
subtypes 1 and 2 (hypocellular) were compared with subtypes
3 and 4 (hypercellular). The results are shown in Table 3. An
association between hypocellularity and a low capsular thick-
ness was con®rmed for parotid tumours only where the total
capsular readings were analysed.
138 A.J. Webb & J.W. Eveson

Table 4. Extent of capsular exposure

No. (%)

Total Neoplasms 126

Enucleation±local excision 18y
Morcellation 4
Standard surgical procedures
Total 104
Superficial parotidectomy 89
Submandibular gland excision 15
Capsular exposure  50% 64 (61.5%)
`Intact' tumours 21 (20%)z

Gross damage 7/126 (5.5%).
yGross damage 3/18 (16.6%).
zEighteen parotid, three submandibular.

Figure 9. Cellular pleomorphic adenoma (`adenoid cystic'), Seifert

type 4. Haemotoxylin and eosin, 25. enucleation±local excision resulted in 100% capsular expo-
sure. The results are shown in Table 4. Gross surgical damage
was present in 7/126 (5.5%) tumours. In respect of neoplasms
treated by `standard excisions', 81% revealed some degree of
capsular exposure and in 61.5% this exceeded 50% of the
discernible surface.

Figure 10. Seifert type 2 (hypocellular) tumour. Haemotoxylin and
eosin, 6.
The extent of capsular exposure was measured from the
histological sections. A 1-mm enclosure by salivary gland or
other tissue (fascia) was considered adequate. In addition, an
assessment was possible from the sections as to whether the
surgical procedure was enucleation±local excision, paroti-
dectomy or morcellation (`piecemeal' removal). By de®nition,
g r a di ng
Grading into histopathological subtypes was attempted for the
126 pleomorphic adenomas (Fig. 11). In three, variability in
tumour pattern obviated a con®dent reading. For the remain-
der (123/126, 97.6%), microscopy classi®ed subtypes as
shown in Table 5.
All 19 submandibular pleomorphic adenomas were graded
and the prevalent impression that, at this site, tumours are
more `cellular' than those in the parotid, was con®rmed: 8/19
(42%) were Seifeit subtype 3 (hypercellular), whereas the
equivalent for subtypes 3 and 4 combined, in the parotid
lesions, was 27/104 (25.9%).

Table 3. Tumour cellularity and capsular thickness in submandib-

ular and parotid pleomorphic adenomas

Gland P

Submandibular
Maximal dimensions of capsular thickness (mm) 0.352
Total dimensions of capsular thickness (mm) 0.55
Parotid
Maximal dimensions of capsular thickness (mm) 0.22
Total dimensions of capsular thickness (mm) < 0.001y

Not significant; ysignificant. Figure 11. Seifert type 1 tumour. Haemotoxylin and eosin, 25.

# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142

 Pleomorphic adnenomas of the major salivary glands 139
Table 5. Pleomorphic adenoma subtypes: Seifert classification
Stromal component No.
Subtype 1 30% 50% 45 (Subm. 7)
(Par 38)
Subtype 2 50% 80% 43 (Subm. 4)
(Par 39)
Subtype 3 20% 30% 33 (Subm. 8)
(Par 25)
Subtype 4 20% 30% 2 (Par)
(Adenoid cystic pattern)
Total 123
Subm. ˆ submaudibular; par ˆ parotid.
Discussion
There are few studies on capsular status in major gland
pleomorphic adenomas. Lawson44 investigated tumour size
and capsule in seven parotid lesions and concluded that the
deep surface possessed a thicker capsule and less `bud'
formation than elsewhere. Naeim,45 in a detailed study of
112 tumours, had enunciated three features of capsular thick-
ness in relation to tumour pattern, namely:
1. Hypercellularity increases capsular thickness.
2. Hypocellularity is associated with focal capsular absence
in 69% of cases. For hypercellular tumours, the equivalent
was 22%.
3. Tumour overall size is related to capsular thickness; small
tumours possessed thicker capsules.
A time sequence was proposed for the evolution of pleo-
morphic adenomas: a small hypercellular adenoma growing
into a larger hypocellular entity with sites of capsular absence,
microinvasion and `bud' formation. Our data did not include
`alleged duration' of the tumour. The most precise study
undertaken46 involved 15 parotid lesions using whole organ
sections, to permit a three-dimensional assessment of tumour
shape, capsular exposure and surgical damage. A signi®cant
conclusion was that cellularity bore no relation to capsular
thickness. These opposed views added signi®cance to our
investigations, especially as we found minimal correlation of
capsular morphometry with cellular features. The Seifert
grading into subtypes was both feasible and reproducible
for 97.3% of pleomorphic adenomas (Figs 9±11). Comparing
parotid and submandibular adenomas in respect of capsule, we
found that the former showed thicker capsules (P < 0.001) but
the latter were more cellular.
Comparing 20 of the smallest with 20 of the largest
adenomas, we con®rmed that small tumours tended to be
hypercellular and that for the parotid alone hypocellular
tumours were associated with low capsular thickness
(P < 0.001). In practical terms this is a signi®cant ®nding
and supports the widely-held view that large (i.e. >25 mm
diameter) parotid adenomas can be a problem at operation.
# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142
Present series Seifert et al. [47]
37.6 37
34.9 53.5
26.8 9
1.6 6.5
They possess thinner capsules, abundant myxoid stroma
(Seifert subtype 2) and are inclined to rupture during dissec-
tion.41 This hazard is extended by our ®nding that focal
capsular absence was detected in 74% of 45 Seifert subtype
2 tumours.
The phenomenon of capsular exposure at surgery was
de®ned by Donovan and Conley,15 who estimated that in
50% of their parotidectomy specimens, close capsular dissec-
tion was unavoidable. Of 15 pleomorphic adenomas, all had
revealed some exposure and in four (27%) surgical capsular
injury was evident microscopically.
Our ®ndings from 126 primary adenomas (1984±1995)
treated surgically are both con®rmatory and disturbing. Enu-
cleation of parotid tumours was performed in 19 and mor-
cellation (piecemeal removal) in four. In these, there was
100% capsular exposure and in 3/19 enucleations (15.7%)
gross surgical damage was detected.
Standard surgical procedures (110) were performed in
accordance with accepted operative techniques, yet in 64/
110 (58.1%) there was extensive ( 50%) capsular exposure:
in only 21/110 (19%) was complete enclosure of the adenoma
anticipated. This proportion could be an overestimate using
our methodology and a ®gure of 100% exposure at some site
might be more realistic. There is no consensus in the literature
as to a satisfactory margin for pleomorphic adenoma excision.
General statements such as `remove with a margin that looks
normal to the naked eye' are commonplace.47±49 Hancock,33
an enthusiast for extracapsular enucleation, described a `small
margin or fringe of normal tissue'; yet for 64 primary parotid
tumours he reported 40% with a covering; 38% part exposed;
and 22% totally exposed. Various publications15,21,46,50 sug-
gest a range of 1±10 mm as adequate and McGurk10 observed
that 50% of parotid adenomas are at some point, apposed to
seventh nerve divisions. Our microscopical measurements,
taking into account capsular microinvasion and bud forma-
tion, suggested that 1 mm of normal tissue was acceptable.
In support of the feasibility of extracapsular dissection or
enucleation the pericapsular areolar space is emphasized and
illustrated;32,33,51 but our ®nding of a limited space in only
140 A.J. Webb & J.W. Eveson
28% of slides examined suggests that arti®cial cleavage may
arise during attempted enucleation (Figs 4±6).
A question mark remains, however, as descriptions of
operative technique are confusing.33,51 In addition, advocates
for super®cial parotidectomy place little emphasis on the
extent of capsular exposure encountered during the opera-
tion,26,47,52 yet this operation fails to ensure tissue enclosure of
the tumour. Possible evidence for tumour growth character-
istics was important.11,22,23,53 Surface bosselation and general
irregularity receives mixed emphasis. Lam46 outlined a profu-
sion of surface distortions and our observations were that in
58.4% of pleomorphic adenomas, bosselae were seen, while in
20.4% these protrusions were extreme. Bosselation must be a
manifestation of neoplastic growth pattern and we regarded
the associated `capsular enveloping' phenomenon as a con-
®rmation (Fig. 5). This feature was seen in 33% of adenomas.
Epithelial margination indicates a neoplastic growth advan-
cing edge pattern and is particularly conspicuous in myxoid
bosselae (Fig. 6).
Microinvasion of the capsule is well described in pleo-
morphic adenomas and together with tumour buds is an
indication of neoplastic activity (Fig. 7). We found tumour
buds in 15/126 (11.9%). All `buds' were bounded by a thin
®brous capsule and were closely connected to the main
tumour mass (Fig. 1). We found no evidence of multicentricity
in the 126 primary tumours. Staining for reticulin enhanced
the impression from haemotoxylin and eosin slides that the
pleomorphic adenoma capsule can be strikingly lamellated
(Fig. 12). The reticulin ®bres align in the direction of the
lamellae. This feature, along with bosselation and enveloping,
lead ustospeculatewhether the capsuleis a product ofthe tumour
itself and not the host tissue as is usually stated.11±13,41,43
The concern in respect of FNA perversion of tumour
morphology was not supported: of 126 tumours, only 8%
Figure 12. Pleomorphic adenoma capsule showing fibrous lamella-
tion and pericapsular space. Capsular thickness and space 75 mm.
Reticulin stain  40.
# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142
showed any possible evidence of a needle track (Fig. 8).
Infarction was a minor feature (4/126, 3.2%) and in these
cases coexisting tracks were not found. We concluded that
carefully performed FNA (needle size 21±25 gauge) should
not prejudice a standard histopathological diagnosis.
Recurrent pleomorphic adenomas, continue to be a serious
clinical problem, but the precise causation is controver-
sial.37,54 A rate of 1% 2% over a 10-year interval can be
achieved but, with notable exceptions,32,33,39,52 general clin-
ical follow-up is poor and anecdotal evidence suggests that
recurrence rates may be higher than the best reported ®gures.
Despite every care, tumour rupture can complicate sur-
gery.30,31,55 It has been proposed that single recurrences,
which constitute the minority, derive from a residual isolated
bud, whereas multiple ®eld recurrence results from tumour
rupture.56 The revelation of tumour buds led to strong con-
demnation of surgical enucleation in any guise.23 Neverthe-
less, the proponents of extracapsular dissection have rejected
tumour `buds' as a cause of recurrence and invoke surgical
inadequacy. This matter is unresolved. Many surgeons report
rates of tumour rupture and minor capsular damage in the
range of 9% 50%.28±31,52,57 Some assert that spillage of cells,
inadequate microscopical margins and known residual tumour
foci can be effectively controlled by external radiotherapy, but
this view is also strongly contested.33,51,58,59
Another reason advanced for recurrence is the intrinsic
neoplastic biology of the tumour,60,61 as such patients tend to
be signi®cantly younger than the mean age for primary
pleomorphic adenoma presentation. Also parotid surgery in
the young±especially females±might be suboptimal for fear of
facial nerve damage.26,32,33,54,62
A very important aspect of parotid pleomorphic adenoma
surgery is the action required if there is accidental rupture. The
question of irrigation arises and this subject receives minimal
attention in published work. The use of 50% zinc chloride
solution; Harrington's solution (mercuric chloride±hydrochlo-
ric acid±ethyl alcohol), iodine and 60% ethyl alcohol appears
in historical literature.1,24,63 More recent reports favour a
variety of agents: 0.15% Cetrimide, normal saline, and sterile
distilled water alternating with saline.37,45,64,65 Common
knowledge suggests that the practice of wound irrigation is
rare. Our preference, in the event of tumour rupture, is for
liberal sterile water irrigation acting as a tumouricidal agent
followed by normal saline (200±500 ml) as repeated 20 ml
syringe washouts. Reported rates for temporary facial neur-
apraxia following parotidectomy or extracapsular dissection
range from 11% to 47%.32,51,64,66 Our observations are anec-
dotal over 27 years, but it seems possible that peroperative
saline irrigation reduces facial nerve dysfunction and
diminishes the risk from minor capsular breaches leading
to tumour seeding.
Is there common ground between the seemingly diametri-
cally opposed surgical procedures? Fervent extracapular
 Pleomorphic adnenomas of the major salivary glands 141
dissectors do regard this operation as suitable only for the very
expert or `surgical connoisseur',32,33,51 and that in routine
surgical practice some form of super®cial parotidectomy is the
safer option,17,51 It is important to re-emphasize, that for
parotid pleomorphic adenomas so treated, only 12±40% will
emerge with an adequate enclosure by parotid tissue, fascia or
platysma. A majority of histopathological reports in our series
questioned the completeness of excision.
For submandibular pleomorphic adenomas, enucleation is
particularly hazardous in respect of recurrence and there is a
consensus that total submandibular gland excision with the
adenoma is essential. In situations of diagnostic uncertainty,
some surgeons add a suprahyoid node dissection and soft
tissue clearance.67
Salivary gland surgeons must be prepared for very precise
capsular dissection, especially where the facial nerve is
apposed, to avoid capsular rupture and neural injury. Even
so, accidental injury may occur, but Natvig and Soberg55
opined that, if properly managed, this will not lead to recur-
rence and even a macroscopically clear and microscopically
con®rmed excision may carry a small recurrence rate. Our
studies have provided ample evidence for the ¯imsy, variable,
and uncertain border between a pleomorphic adenoma and the
host tissue. The ®ndings suggest that parotidectomy of whatever
form should guarantee at least some adequate tissue margin
around the tumour. For some surgeons, partial super®cial par-
otidectomy has evolved as an ideal procedure,60,65 but others will
prefer more extensive conservative parotidectomy to maximize
safety in both excision and nerve preservation whilst also
providing a clear view of the whole operative ®eld.52,54,64,66
References
1 SISTRUNK W.E. (1921) Tumors of the parotid gland: Mayo Clinic
No Clinic of Walter E. Sistrunk. Surg. Clin. N. Am. 1, 1515±1521
2 ADSON A.W. & OTT W.O. (1923) Preservation of the facial nerve in
the radical treatment of parotid tumors. Arch. Surg. 6, 739±746
3 PATEY D.H. (1930) The mixed tumours of the salivary glands.
BJS 18, 241±245
4 PATEY D.H. & HAND B.H. (1952) Diagnosis of mixed parotid
tumours. Lancet 2, 310±311
5 WEBB A.J. (1973) Cytologic diagnosis of salivary gland lesions
in adult and pediatric surgical patients. Acta Cytol. 17, 51±58
6 FITZGIBBON G.M. (1953) Discussion: surgical treatment of
parotid tumours. Proc. Roy. Soc. Med. 46, 1016±1018
7 EDVALL C.A. (1954) Mucous and salivary gland tumours. Acta
Chir. Scand. 107, 313±320
8 PATEY D.H. (1954) The present position of parotidectomy in the
surgery of the parotid gland. Arch. Middx Hosp. 4, 91±105
9 GLEAVE E.N., WHITTAKER J.S. & NICHOLSON A. (1979) Salivary
tumours: experience over thirty years. Clin. Otolaryngol. 4,
247±257
10 MCGURK M. (1997) Parotid pleomorphic adenoma. Review.
BJS 84, 1491±1492
11 ENEROTH C.-M. (1965) Mixed tumors of major salivary glands,
prognostic role of capsular structure. Ann. Otol. Rhinol.
Laryngol. 74, 944±953
# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142
12 PERZIK S.L. (1956) The case against enucleation. Calif. Med.
85, 27±29
13 EVANS R.W. & CRUICKSHANK A.H. (1970) Epithelial tumours
of the salivary glands. In Problems in Pathology, pp. 167±225.
W.B. Saunders Co., Philadelphia
14 THACKRAY A.C. & LUCAS R.B. (1974) Tumors of the major
salivary glands. In Atlas of Tumor Pathology. Second Series,
Fascicle 10. Armed Forces Institute of Pathology, Washington
DC
15 DONOVAN D.T. & CONLEY J.J. (1984) Capsular significance in
parotid tumor surgery: reality and myths of lateral lobectomy.
Laryngoscope 94, 324±329
16 EVESON J.W. & CAWSON R.A. (1985) Salivary gland tumours: a
review of 2410 cases with particular reference to histological
types, site, age and sex distribution. J. Pathol. 146, 51±58
17 MCGURK M., RENEHAN A., GLEAVE E.N. et al. (1996) Clinical
significance of the tumour capsule in the treatment of parotid
pleomorphic adenomas. BJS 83, 1747±1749
18 MASSON P. (1924) Tumeurs des glandes annexes des muqueses
de la face et du cou. In Atlas Du Cancer III et IV. BallieÁre, Paris.
19 AHLBOM H.E. (1935) Mucous and salivary gland tumours. A
clinical study with special reference to radiotherapy: based
upon 254 cases treated at Radium Hemmet Stockholm. Acta
Radiol. 23(Suppl.), 1±452
20 DOCKERTY M.B. & MAYO C.W. (1942) Primary tumors of the
submaxillary gland with special reference to mixed tumors.
Surg. Gynae. Obstet. 74, 1033±1045
21 JANES R.M. (1957) Tumours of the parotid gland. Ann. Roy.
Coll. Surg. Eng. 21, 1±20
22 WILLIS R.A. (1960) Epithelial tumours of salivary glands. In
Pathology of Tumours 3rd edn, pp. 321±349. Butterworths,
London
23 PATEY D.H. & THACKRAY A.C. (1958) The treatment of parotid
tumours in the light of a pathological study of parotidectomy
material. BJS 45, 477±487
24 BAILEY H. (1941) The treatment of tumours of the parotid gland
with special reference to total parotidectomy. BJS 28, 237±246
25 REDON H. (1953) Discussion: the surgical treatment of parotid
tumours. Proc. Roy. Soc. Med. 46, 1014±1016
26 MAYNARD J.D. (1967) Parotid enlargement. Br. J. Hosp. Med. 7,
620±624
27 HOBSLEY M. (1973) Salivary tumours. Br. J. Hosp. Med. 10,
553±562
28 SMIDDY F.G. (1956) Treatment of mixed parotid tumours. Br.
Med. J. 1, 322±325
29 HAW C.S. (1975) Pleomorphic adenoma of the parotid gland. J.
Roy. Coll. Surg. Edin. 20, 25±29
30 MCEVEDY B.V. & ROSS W.M. (1976) The treatment of mixed
parotid tumours by enucleation and radiotherapy. BJS 63,
341±342
31 ARMITSTEAD P.R., SMIDDY F.G. & FRANK H.G. (1979) Single
enucleation and radiotherapy in the treatment of pleomorphic
salivary adenoma of the parotid gland. BJS 66, 716±717
32 GUNN A. & PARROT N.R. (1988) Parotid tumours: a review of
parotid surgery in the Northern Regional Health Authority of
the United Kingdom 1978±1982. BJS 75, 1144±1146
33 HANCOCK B.D. (1987) Pleomorphic adenomas of the parotid:
removal without rupture. Ann. Roy. Coll. Surg. Eng. 69,
293±295
34 HYBBINETTE S. (1936) Contribution a l'etude des tumeurs
beÂnigues de la parotide et a leur traitement radio: chirurgique.
Acta Chir. Scand. 77, 19±49
35 FOOTE F.W. Jr & FRAZELL E.L. (1953) Tumors of the major
salivary glands. Cancer 6, 1065±1113
36 BEAHRS O.H. & ADSON M.A. (1958) The surgical anatomy and
technique of parotidectomy. Am. J. Surg. 95, 885±896
142 A.J. Webb & J.W. Eveson
37 NORMAN J.E. & B. (1995) Recurrent mixed tumors of the
major and minor salivary glands. In Color Atlas and Text of
the Salivary Glands: Diseases, Disorders and Surgery, pp.
229±242. Mosby Wolf, London
38 WEBB A.J. (1982) Surgical aspects of aspiration biopsy
cytology. In Recent Advances in Surgery: No. 11, pp. 39±67.
Churchill Livingstone, Edinburgh
39 HUGHES R.G. & LYONS T.J. (1982) Parotid tumours, diagnosis
and treatment. Lancet 1, 1080
40 CONLEY J., THAWLEY S.E., WARD S.P. et al. (1974) Discussion:
basal cell adenoma of the salivary glands. Laryngoscope 84,
1756±1765
41 ELLIS G.L. & AUCLAIR P.L. (1996) Tumors of the salivary
glands. In Atlas of Tumor Pathology. Second Series, Fascicle
12, pp. 441±467. Armed Forces Institute of Pathology,
Bethesda, MA
42 HAJDU S. & MELAMED M.R. (1984) Limitations of aspiration
cytology in the diagnosis of primary neoplasms. Acta Cytol. 28,
337±345
43 SEIFERT G., LANGROCK I. & DONATH K. (1976) Pathomorpholo-
gische Subklassifikation der pleomorphen SpeicheldruÈsen
adenomen. HNO 24, 415±426
44 LAWSON H.H. (1989) Capsular penetration and perforation in
pleomorphic adenoma of the parotid salivary gland. BJS 76,
594±596
45 NAEIM F., FORSBERG M.-I. Jr, WAISMAN J. et al. (1976) Mixed
tumours of the salivary glands: growth pattern and recurrence.
Arch. Pathol. Lab. Med. 100, 271±275
46 LAM K.H., WEI W.H., HO H.C. et al. (1990) Whole organ
sectioning of mixed parotid tumors. Am. J. Surg. 160, 377±381
47 PATEY D.H. (1953) Discussion: surgical treatment of parotid
tumours. Proc. Roy. Soc. Med. 46, 1014±1020
48 DYKUN R.J., DEITEL M., BOROWY Z.J. et al. (1980) Treatment of
parotid neoplasms. Can. J. Surg. 23, 14±19
49 HOBSLEY M. (1986) Swellings in the parotid region. In
Pathways in Surgical Management, pp. 46±58. Edward Arnold,
London
50 YAMASHITA T., TOMODA K. & KUMAZAWA T. (1993) The
usefulness of partial parotidectomy for benign parotid gland
tumors: a retrospective study of 306 cases. Acta Otolaryngol.
(Stockh.) 500(Suppl.), 113±116
51 GLEAVE E.N. (1995) An alternative to superficial parotidect-
omy: extracapsular dissection. In Color Atlas and Text of
the Salivary Glands: Diseases, Disorders and Surgery, pp.
165±172. Mosby Wolf, London
52 STEVENS K.L. & HOBSLEY M. (1982) The treatment of
pleomorphic adenomas by formal parotidectomy. BJS 69, 1±3
53 ROSAI J. (1989) Major and minor salivary glands. In Ackerman's
Surgical Pathology, 8th edn, pp. 815±856. CV Mosby
Company, St Louis
54 MARAN A.G.P., MACKENZIE I.J. & STANLEY R.F. (1984)
Recurrent pleomorphic adenoma of the parotid gland. Arch.
Otolaryngol. 110, 167±171
# 2001 Blackwell Science Ltd, Clinical Otolaryngology, 26, 134±142
55 NATVIG K. & SOBERG R. (1994) Relationship of intraoperative
rupture of pleomorphic adenomas to recurrence: an 11±25 year
follow-up study. Head Neck 16, 213±217
56 RENEHAN A., GLEAVE E.N., HANCOCK B.D. et al. (1996) Long-
term follow up of over 1000 patients with salivary tumours
treated in a single centre. BJS 83, 1750±1754
57 BUCHMAN C., STRINGER S.P., MENDENHALL W.M. et al. (1994)
Pleomorphic adenoma: effect of tumor spill and inadequate
resection on tumor recurrence. Larygoscope 104, 1231±1234
58 SHAHEEN O.H. (1989) Total parotidectomy: a routine treatment
for parotid gland swellings? J. Laryngol. Otol. 103, 181±186
59 WATKIN G.T. & HOBSLEY M. (1986) Influence of local surgery
and radiotherapy on the natural history of pleomorphic
adenomas. BJS 73, 74±76
60 LIU F.F., ROTSTEIN L., DAWSON A.J. et al. (1995) Benign parotid
adenomas: a review of the Princess Margaret Hospital
experience. Head Neck 17, 173±183
61 NIPARKO J.K., BEAUCHAMP M.L., KRAUSE C.J. et al. (1986)
Surgical treatment of recurrent pleomorphic adenoma of
the parotid gland. Arch. Otolaryngol. Head Neck Surg. 112,
1180±1184
62 RENEHAN A., GLEAVE E.N. & MCGURK M. (1996) An analysis of
the treatment of 114 patients with recurrent pleomorphic
adenomas of the parotid gland. Am. J. Surg. 172, 710±714
63 OCHSNER A.J. (1919) Three cases illustrating certain benign
lesions of the parotid gland: Clinic of Dr A.J. Ochsner. Surg.
Clin. Chicago 1919, 903±915
64 MAYNARD J.D. (1986) Management of pleomorphic adenoma of
the parotid. BJS 75, 305±308
65 LEVERSTEIN H., VAN DER WAL J.E., TIWARI R.M. et al. (1997)
Surgical management of 246 previously untreated pleomorphic
adenomas of the parotid gland. BJS 84, 399±403
66 NORMAN J.E.B., MAYNARD J.D. & GUNN A. (1995) Benign
tumors of the parotid and submandibular glands, management
by sialadenectomy. In Color Atlas and Text of the Salivary
Gland: Diseases, Disorders and Surgery, pp. 139±164. Mosby
Wolf, London
67 BRADLEY P.J. (1999) Submandibular gland and minor salivary
gland neoplasms. Curr. Opin. Otolaryngol. Head Neck Surg. 7,
72±78
68 ENEROTH C.-M. & ZAJICEK J. (1966) Aspiration biopsy of
salivary gland tumors. III: Morphologic studies on smears and
histologic sections from 368 mixed tumours. Acta. Cytol. 10,
440±454
69 ORELL S.R. & NETTLE W.J.S. (1988) Fine needle aspiration
biopsy of salivary gland tumours: problems and pitfalls.
Pathology 20, 332±337
70 HOBSLEY M. (1981) Sir Gordon Gordon-Taylor: two themes
illustrated by the surgery of the parotid gland. Ann. Roy. Coll.
Surg. Eng. 63, 198±202
71 MCGURK M. & HUSSAIN K. (1997) Role of fine needle aspiration
cytology in the management of the discrete parotid lump. Ann.
Roy. Coll. Surg. Eng. 79, 198±202