“A square peg into a round hole”: a modified rhomboid flap

and its clinical application

A. A. QUABA and B. C. SOMMERLAD
North East Thames Regional Plastic Surgery Unit, St Andrew’s Hospital, Billericay and The London Hospital

Summary---Two modifications to the classic design of the rhomboid flap are described. Firstly, no
attempt is made to engineer a rhomboid defect and, secondly, the flap is made smaller than the defect
to be reconstructed. The advantages of these modifications are discussed and their application in 175
reconstructions is reported.

Clinical applications of a modified rhomboid flap equal to the extension, drawn at 60 to it, and is
almost parallel to one side of the defect, following
Alexander Limberg (1946, 1966) described the use
its curve, but care should be taken not to narrow
of a fap to fill a rhomboid-shaped defect. Lister
the base of the flap (Fig. l), The flap is raised and
and Gibson (1971) amplified the geometry and de-
the secondary defect closed directly after under-
sign of the rhomboid flap, emphasising the classic
mining if necessary. The flap is then positioned into
measurements: all angles being 60 or I20 and all
the defect with kev stitches to distribute the tension
sides being equal. A number of modifications have
evenly. Although-it may look rather like putting a
been described (Dufourmentel. 1963; Webster et
“square peg into a round hole”. surprisingly. it is
trl.. 1978; Becker, 1979). Most of the published
rarely necessary to trim the corners of the flap.
work concentrated on the mathematical principles
When suturing is completed, the configuration of
of the flap rather than on its clinical application
the sutured wound would not be significantly dif-
(Jervis rt trl.. 1974; Borges. 198 1; Larrabee et (11..
ferent from that which follows the classic operation
1981).
(Fig. 2).
This paper describes a clinical approach to the
use of the rhomboid flap with two modifications.

Modifications to the classic design

The lesion is excised as necessary. without consider-
ing the shape of the defect produced; corners need
not be sacrificed to produce a rhomboid-shaped de-
fect. Many defects will end up having an almost cir-
cular shape. The “diagonal” of the defect, to be
extended as one side of the flap, is then chosen. It
may not be the shortest but may be selected so that
the donor site can be closed most readily and with
the lea.st conspicuous scar. Choosing the shortest
diagonal. however. does make flap transposition
easier and minimises dog-ear formation. These fac- & = & = A& = Cb
AB = BC = AC = >,!3 CD
tors have to be weighed up in each case.
The flap is always planned to be smaller than the MODIFIED DESIGN CLASSIC DESIGN
defect. The chosen “diagonal” is. therefore, Fig. I
extended by about two-thirds of its own length; the Figure I-Comparison between the modified and the classic de-
exact length depends on the tension of the skin sur- signs. The defect LS not necessarily rhomboid-shaped and the
rounding the defect. The other side of the flap is flap is smaller than the defect.
 BRITISH JOURNAL OF PLASTIC SURGERY

Fig. 2
Figure 2-Fitting a “square peg into a round hole”. In the modified design the positioning of the flap into the defect is not predeter-
mined. This aliows for more flexibility and better distribution of tension.

Clinical application were equally represented. Table I shows the patho-

logical processes requiring excision and Table 2
A hundred-and-seventy-five patients who had re-
and (Fig. 3) show the anatomical distribution of
constructions using the modified flap were
the resulting defects. One-third of the reconstruc-
reviewed.
tions were performed by a consultant and two-
The average age was 59.8 years and the two sexes
thirds by trainees. About half of the operations
were carried out under local anaesthesia.
Table 1 Pathological processes requiring excision

Results
Basal cell carcinoma 113 (64.6)
Squamous cell carcinoma 17 (9.7) Post-operative complications were few (Table 3).
Malignant melano~ 16 (9.2) Partial flap necrosis, with or without partial wound
Others 29(!6.5) dehiscence, was noted in eight patients (4.5%). It
usually followed closure under escessive tension,
TOGIl 175 (100)
particularly in the forehead and scalp region, or
when the base of the flap was close to a fixed land-
mark such as the preauricu~ar margin. In no case
Table 2 Anatomical distribution of skin defects was necrosis sufficient to warrant further surgery.
The cosmetic result, assessed at an average
follow-up interval of 1.7 years, was judged with
Forehead and scalp 31 (12.0)
Temporal area 23(1X1)
Outer and inner canthi 37(21.2) Table 3 ColnpIications
Cheeks 31 (17.7)
Nose 14 (8.0)
Upper lip and commissures 14 (8.0)
Lower lip and chin 5 (2.9) Infection 6 (3.5)
Post-auricular and mastoid 12 (6.8) Haematoma 2(1.1)
Others 18 (10.3) Partial necrosis and/or partial dehiscence 8 (4.5)

Total 175 (100) Total 16(9.1)

“A SQIIARE PIIGINTOA ROUNDHOLE" 165

particular reference to scarring, distortion of the

landmarks, trapdooring and dog-ear formation.
Most of the scars were classified as hairline and
some were hardly noticeable (Fig. 4). In 9% of
cases the scar was significantly stretched. especially
at the site ofclosure of the secondary defect. A few
of the scars. especially those on the trunk. were
hypertrophic (4.5%).
There was no distortion of facial landmarks in
more than three-quarters (77.5%) of the patients.
Some distortion (mainly of the hairline, srdeburn
and eyebrow) was noted in 22.5% of the cases
reviewed. This caused minimal cosmetic but no
functional deficit.
Trapdooring (pin-cushioning) was observed in
9O/0of patients, especially where small defects were
closed with apparently over-generous flaps.
Dog-ears were visible in 22% of cases and were
always at the hinge-point of the base of the flap and
never at the apex of the donor site.

Discussion
There are three advantages of not engineering a
rhomboid defect:
The positioning of the transposed flap into the
defect is not predetermined. it can be sited to
distribl~te tension evenly. The geometric and
paper models of the classic design assume
rigidity of size and shape of the skin defect and
of the flap and that. of course. is not true in the
clinical situation.
There is more flexibility in choosing the donor
site. The classic design allows only four possible
choices of flaps while our modification gives an
unlimited choice of flaps (Fig. 5) allowing the
scar to be placed in the least conspicuous site
and minimising distortion of landmarks.
Tissue does not need to be unnecessarily sacri-
ficed to produce a rhomboid defect.
The second modification, where the flap is made
smaller than the defect and the surrounding tissues
contribute to the closure, allows the donor site to
be closed more easily. This means that the flap can
be used to reconstruct larger defects than would be
possible in the classical description. In addition, the
transmission of even tension across the flap may
contribute to the low incidence of trapdooring, as
was noticed in this series.
A study of the pattern of scar stretching and the
distortion of landmarks suggests that. following
transposition, the area of maximum tension is
 Fig. 4
Fig’ure &-Two operative series and post-operative results to demonstrate the use of the modified Hap.
4 syUARE PEG INTO A ROUND HOLE” 167

Fig. 4
 BRITISH JOURNAL OF PLASTIC SURGERY

l. / Fig. 6
‘\

Fig. 5 Figure &--Elliptical excision and direct closure or circular ex-

cision and Rap closure? Better distribution of tension is an argu-
Figure j--After a rhomboid defect has been engineered only ment for the use of the modified design in the closure of skin
four choices of donor site are possible but by keeping the defect defects whew no di~~erenti~i skin laxity exists. A and B are the
circular or nearly so, the options may be unlimited. points of rn~lxirnunl tension but the tension across B is iess than
that wxoss A due to the use oTa flap smaller than the defect and
allowing the surrounding skin to pnrttcipate in the closure.

located across the donor site closure rather than at designed rhomboid Aap (equal short diagonals and
the flap tip. equal skin tension). By using the modified flap,
This modified rhomboid flap is still basically a however, closure of both defects may become
60. transposition flap and one of its great advan- easier due to the smaller size of the donor site and
tages is that the resulting dog-ears are rarely a contribution of the skin surrounding the primary
problem. Even if noticeable at operation, they defect toward its closure, which permits better dis-
settle remarkably well with time and none was tribution of tension (Fig. 6). In this series more
revised in this series. In fact, the good cosmetic re- than 20 fairly large scalp and forehead defects were
sults in these patients have led us to use the rhom- closed by the judicious use of the modified flap.
boid flaps in situations (r.g. the cheeks) where avoiding skin grafting (Figs 3 and 7).
elliptical excision and closure would be possible but Since the completion of our detailed review of
would produce unsightly dog-ears. 175 reconstructions using the modified design, over
There is controversy regarding the use of the 400 patients have had facial and other defects
rhomboid flap in areas where there is no differen- closed using this flap. The results in these patients
tial skin laxity such as the scalp, forehead and confirmed our findings that the flap is versatile, has
back. It may be argued that if there was no differ- few complications and gives superior cosmetic re-
ence of tension in the skin surrounding a defect sults when compared to techniques used pre-
then its direct closure would be no more difficult viously. The flap has become our workhorse for
than closure of the donor site of a classically facial reconstructions.
(\ SQUARE PEG INTO A ROUND HOLE”

Fig. 7
Figure ?--In this patient. ellipttcal excision and closure of the
basal cell carcinoma on his forehead would have resulted tn ex-
cessive tension and upward pulling of the eyebrows. The use of
the modified flap allowed better distribution of tension and en-
abled closure without distortion of landmarks. Note the absence
of dog-ear formation.
170 BRITISH JOURNAL OF PLASTIC SURGERY

References in Plustic Surgery, Second Edition. Edited by Gibson, T.

London: Butterworths.
Becker, F. F. (1979). Rhomboid flap in facial reconstruction. Lister, G. D. and Gibson, T. ( 1972). Closure of rhomboid skin
Archiw r~f’O~okrrygo~og~, 105. 569.
defects: the flaps of Limberg and Dufourmentel. British Jour-
Borges, A. F. (1981). The rhomboid flap. Plus~ic cmd Reconstruct- nal qf‘Piusric Surgery, 25.300.
rive Surgery. 67.458.
Webster, R. C., Davidson, T. M. and Smith, R. C. ( 1978). The
Dufourmentel, C. (1963). An L-shaped flap for lozenge-shaped
thirty degree transposition flap. Laryngoscope. 88,85.
defects. Trunsuc~tiom of’ the Third Imemuriond Congrrss of
P/tr.srk Surgrr~~. p. 722. Amsterdam: Excerpta Medica Foun-
dation.
Jervis, W., Salyer, K. E.. Busquets, M. A. V. and Atkins, R. W.
The Authors
(1974). Further applications of the Limberg and Dufourmen-
tel tlaps. P/us/ic~ tmc/ Rrwn.vtruc/irr Surgrry. 54. 335. A. A. Quaba. FRCS, Senior Registrar in Plastic Surgery, St
Larrabee, W. F., Jr., Tracby, R., Vutton, D. and Cox, K. ( I98 I). Andrew’s Hospital, Billericdy.
Rhomboid flap dynamics. ,4rd~iwx of Ofolqwgdogy. 107. B. C. Sommerlad. FRCS. Consultant Plastic Surgeon. St
755. Andrew’s Hospital, Billericdy and The London Hospital.
Limberg, A. A. ( 1946). Mathwnrr~id principles of IOU/ plustic
.vurgq’ prowdurrs on the surf& of thr hurnan hods. Lenin- Requests for reprints to: Mr A. A. Quaba, FRCS. North East
grad: Medgis. Thames Regional Plastic Surgery Unit. St Andrew’s Hospital.
Limberg, A. A. ( 1966). Design of local Haps. In: Modern Trends Billericay. Essex.