Clinical Note
Clinical Outcome of 285 Medpor Grafts used
for Craniofacial Reconstruction
Roberto Cenzi, MD,* Antonio Farina, MD,y Luca Zuccarino, MD,z Francesco Carinci, MD§
Ferrara, Italy
Porous polyethylene (Medpor) is an alloplastic material worldwide used for craniofacial reconstruction. To evaluate complications and risk factors
associated with this synthetic graft, a retrospective
study was performed. A series of 285 Medpor grafts
were placed in 187 patients. Age, sex, diagnosis at
admission, site, type of surgical insertion, type of
fixation, and outcome (no complications, anesthesia, exposure, infection, and implant remodeling
and removal) are considered. By means of univariate and multivariate analyses, we detect variables
most associated with poor outcome. Univariate
analysis showed that graft ‘‘survival’’ curves stratified according to (1) diagnosis at admission and (2)
site are statistically significant. Subsequently, a Cox
analysis was performed: both variables are also predictors of graft outcome. Porous polyethylene is a
reliable alloplastic material that can be satisfactory
used for craniofacial reconstruction. However, some
sites (i.e., nose, maxilla, and ear) and diagnosis at
admission (i.e., syndromic patients previously operated) are related to an higher risk of implant failure.
operation time. On the contrary, homologue bank tissues and animal derived products are not completely
safe because unknown diseases can be potentially
transferred. Consequently, alloplastic materials are
used for craniofacial reconstruction. Among them,
porus polyethylene (Medpor) has been extensively
used since the 1990s. Its properties make it an excellent choice for correcting cranial and facial defects.
The implant is easy to shape, flexible, remarkably stable, and exhibits rapid soft-tissue ingrowth.
Porus polyethylene has been used to repair
cranial defects,1–4 to restore facial deformities,5–7 to
reconstruct ear8,9 and orbit 10,11, as spherical orbital
prosthesis,12–14 to correct lower eyelid retraction,15,16
and to restore nasal function and shape.17–19 Reported
complications are persisting pain and anesthesia,
implant exposure, infection, and subsequent graft
removal. From this point of view, however, there is
a lack in regards to rate of failure and associated risk
factors. Consequently, we analyzed the clinical outcome of 285 Medpor grafts used for craniofacial reconstruction to detect complications and risk factors.
Key Words: Facial reconstruction, alloplastic material,
Medpor, porous polyethylene, Cox analysis
PATIENTS
C
raniofacial ‘‘skeletal’’ defects should be
ideally correct with autologous bone or
cartilage. However, collecting an adequate
amount of bone from other donor sites of
the same patient is not always possible, it carries
additional morbidity, and it makes for a prolonged
From the *Department of Maxillofacial Surgery, Civil Hospital,
Rovigo, yDepartment of Histology, University of Bologna,
z
Department of Maxillofacial Surgery, University of Ferrara, and
§
Department of Maxillofacial Surgery, University of Ferrara, Italy.
Address correspondence to Dr. Carinci Francesco, Chair of Maxillofacial Surgery, ENT Clinic, Arcispedale S. Anna, Corso Giovecca,
203, 44100 Ferrara, Italy; E-mail: [email protected]
526
AND
METHODS
O
ne hundred eighty-seven patients consecutively
grafted with porous polyethylene for craniofacial reconstruction were enrolled in this retrospective
study. They were operated at the Civil Hospital of
Rovigo between January 1992 and June 1999. The
mean follow-up was 60 months.
Patients
Patients included 103 (55.1%) males and 84 (44.9%)
females. They ranged in age from 5 to 83 (median
age 27) years at the time of admission. The diagnosis
were as follows: 43 (23%) craniofacial fractures, 65
(34.8%) class II and III malocclusions, 35 (18.7%) facial
or craniofacial tumors, and 44 (23.5%) craniofacial malformations (i.e., Treacher-Collins syndrome, Goldenhar syndrome, Binder syndrome, cleft lip and palate,
ear malformations).
OUTCOME OF 285 MEDPOR GRAFTS FOR CRANIOFACIAL RECONSTRUCTION / Cenzi et al
Grafted Sites
Eight sites were grafted. The number of grafts per site
was as follows: 9 ribs, 35 zygomatic bones, 65 cranial
vaults, 56 mandibles, 27 ears, 40 maxillae, 49 orbits,
and 4 noses. Grafts were inserted as follows: 27 inlay,
242 onlay, 10 onlay plus polyethylene fragments,
4 polyethylene fragments, 2 onlay plus inlay. Two
hundred twenty-six (79.3%) grafts were fixed with
screws, whereas the remaining 59 (20.7%) were
sutured.
Treatment and Outcome
Usually, Medpor grafts were inserted in a one-stage
procedure during the treatment of the main pathology. In cancer patients, for example, porous polyethylene was placed in the temporal fossa after tamporalis muscle transposition for maxillary reconstruction; in blow-out fractures, Medpor was used to
restore the continuity of orbital floor; in cases of class
II and III malocclusion as well in syndromic patients,
Medpor was inserted to improve maxillary and mandible contour, to rebuild ear, and to restore nasal
shape and to paranal area. Note that syndromic patients (for example, cleft patients) underwent more
than one surgical procedure so that grafts were inserted very often in scarring tissues. All patients were
followed up and examined on a monthly basis for the
first 6 months after treatment and every 6 months
thereafter. Implant exposure, infection, graft remodeling, and removal were considered treatment failures.
of independent variables such as age and sex and
whether an event (i.e., exposure, infection, and graft
remodeling and removal) is likely occur. If the associated probability was less then 5% (P , 0.05), the difference was considered statistically significant. In
the process of performing the regression analysis,
odds ratios and 95% confidence bounds were calculated. Confidence bounds did not have to include the
value 1.21 Stepwise Cox analysis allowed us to detect
the variables most associated with graft failure.
RESULTS
T
he variables analyzed were age, sex, patient’s
disease, site, type of insertion (i.e., inlay, onlay,
polyethylene fragments, onlay plus polyethylene
fragments, and onlay plus inlay), primary stability
(obtained by screws or sutures), and outcome. The
last was classified as good and failure (if there were
implant exposure, infection, graft remodeling, and
removal).
Global disease-specific survival rate was approximately 93% (Fig 1).
Among the studied variables, we considered
only those associated with a significant effect on graft
survival rate. Survival curves stratified according to
patient diseases are reported in Figure 2. Grafts inserted in syndromic patients have a worse outcome
(x2 =14.87, 3 df, P , 0.0019, log rank test). Figure 3
shows survival curves stratified according to sites:
nose, maxilla, and ear have worse outcomes (x2 = 39,
Statistical Analyses
Univariate analysis
Disease-specific survival curves were calculated
according to the product-limit method (KaplanMeier algorithm) 20. Time zero was defined as the
date of the graft insertion. Grafts that are still inserted
were included in the total number at risk only up
to the time of the patient’s last follow-up. Therefore,
the implant survival rate only changed when one of
the failure variables occurred. The calculated graft
survival rate was the maximum estimate of the true
survival curve. A log rank test was used to compare survival curves, generated by stratifications
for a variable of interest.
Cox regression analysis
Cox regression analysis was then applied to determine the single contribution of covariates on graft
survival rate. Cox regression analysis compares survival data while taking into account the statistical value
Fig 1 Overall disease-specific survival rate.
527
THE JOURNAL OF CRANIOFACIAL SURGERY / VOLUME 16, NUMBER 4 July 2005
7 df, P = 0.0001, log rank test). Survival curves stratified according to type of graft insertion are reported
in Figure 4. No statistically significant difference was
detected (x2 =4.8, 4 df, P , 0.3, log rank test). No difference was noted as regard type of fixation (i.e., screws vs.
sutures) (x2 =2.07, 1 df, P = 0.15, log rank test) (Fig 5).
Cox analysis was then performed by using those
variables that reach a statistically significant value in
univariate analysis (i.e., patient disease and site).
Multivariate analysis was adjusted for patient’s age
and sex (Table 1). Site and type of graft insertion
reached a statistically significant value. Table 2 lists
the series by causes of failure.
DISCUSSION
A
Fig 2 Disease-specific survival rate calculated according
to patient disease categories (x2 = 14.87, 3 df, P , 0.0019,
log rank test). (1) Class II and III malocclusions (96 grafts);
(2) fractures (57 grafts); (3) tumors (38 grafts); (4) malformations (93 grafts).
Fig 3 Disease-specific survival rate calculated according
to grafted site categories (x2 = 39, 7df, P = 0.0001, log rank
test). (1) Rib (9 grafts); (2) zygomatic bone (35 grafts); (3)
cranial vault (65 grafts); (4) mandible (56 grafts); (5) ear
(27 grafts); (6) maxilla (40 grafts); (7) orbit (49 grafts); (8)
nose (4 grafts).
528
lthough autologous bone and cartilage graft are
the gold standard for craniofacial reconstruction, they carry additional morbidity related to the
second operation field and to a prolonged operation
time. Moreover, bone grafts resorb in a way that is not
predictable, and in some sites (such as the temporal
fossa), they can not be used. Consequently, alloplastic
materials have a specific role in craniofacial reconstruction, especially because they are more safe compared with homologue bank tissues and animal
derived products, which can potentially transfer
diseases of an unknown etiology.
Fig 4 Disease-specific survival rate calculated according
to type of graft insertion (x2 = 4.8, 4df, P , 0.31, log rank
test). (1) Inlay (n = 27); (2) onlay + polyethylene fragments
(n = 10); (3) onlay (n = 242); (4) polyethylene fragments (n = 4);
(5) onlay + inlay (n = 2).
OUTCOME OF 285 MEDPOR GRAFTS FOR CRANIOFACIAL RECONSTRUCTION / Cenzi et al
Table 1. Cox Regression Analyses Performed by
Considering Univariate Significant Variables
95.0% Confidence
Interval for Hazard Rate
Variables
Hazard Rate
Lower
Upper
P Value
.79
.986
.295
.95
2.112
1.024
ns
ns
1.367
2.075
1.032
1.218
1.811
3.536
.029
.007
Sex
Age
Graft site
Diagnosis
Medpor is a pure polyethylene with a unique manufacturing process and pore size. Technically, it is easy
to work with; it can be carved, contoured, adapted, and
fixated to obtain a precise, three-dimensional construct.
Physically, it is a pure, biocompatible, and strong substance that does not resorb or degenerate. It demonstrates long-term stability, high tensile strength, and
a virtual lack of surrounding soft-tissue reaction.
Several articles have shown porus polyethylene’s effectiveness to restore craniofacial defects.1–19
Few complications are reported: persistent pain, paresthesia, implant exposure, infection, and subsequent
graft removal. From this point of view, however, there
is a lack in regards to rate of failure and associated
risk factors, and therefore we analyzed the clinical
outcome of 285 Medpor grafts to detect variables
associated with implant failure.
Our global rate of complication is 6.31%,
which is comparable with that reported in large
series.5–7,11,13,16 Main causes of failure are exposure
with subsequent infection. Implant removal was
the treatment in most cases. Sometimes, it was possible to remodel the graft to remove the infected part of
the graft (Table 2). Diagnosis at admission and site are
associated with a statistically significant higher risk
of failure (Table 1).
As regards the site, nose, maxillae, and ear are
sites with worse outcome. Previous studies in large
series17,18 have shown that porous polyethylene implants can be successfully used for nasal reconstruction, but infections are possible because grafts are
usually inserted in secondary rhinoplasty where extensive scarring and thin, soft tissue coverage are encountered. Ear reconstruction has similar problems,
especially those related to a thin skin. Moreover, different areas of a single site have different outcomes.
In the maxilla, for example, grafts placed in the paranasal area have an higher risk of infection compared
with those inserted in other part of upper jaw and in
the ear; grafts inserted behind the cartilage to increase
the outer projection have a better prognosis compared with those used for ear reconstruction.
Seven of 40 grafts placed in the maxilla have mucosal exposure and then infection. Most complications arose in syndromic patients where notable
scars of previous operations were present. Those
scars reduced tissue elasticity. The same factor explains why diagnosis at admission has a significant
statistical impact. In fact, syndromic patients (for example, cleft patients) underwent more than one surgical procedure so that grafts were inserted very often
in scarring tissues.
Table 2. Distribution of the Series According to Causes of Failure
Sex
Age
F
18
F
F
26
23
F
F
Diagnosis
Site
Fixation
Type of Insertion
Cause of Failure
Fracture
Maxilla
Screws
Onlay
Removal
Malformation
Malformation
Maxilla
Maxilla
Screws
Screws
Onlay
Onlay
Removal
Removal
17
6
Malformation
Malformation
Maxilla
Maxilla
Screws
Screws
Onlay
Onlay
Removal
Removal
F
M
6
30
Malformation
Malocclusion
Maxilla
Mandible
Screws
Screws
Onlay
Onlay + fragments
Removal
Removal
M
M
29
25
Fracture
Fracture
Zygomatic bone
Mandible
Screws
Screws
Onlay
Onlay
Removal
Removal
M
M
8
8
Malformation
Malformation
Ear
Ear
Sutures
Sutures
Onlay
Onlay
Removal
Removal
M
7
Malformation
Nose
Sutures
Onlay
Removal
M
M
7
33
Malformation
Malformation
Nose
Ear
Sutures
Sutures
Onlay
Onlay
Removal
Removal
M
M
43
27
Malformation
Malformation
Mandible
Mandible
Sutures
Screws
fragments
Onlay
Removal
Removal
M
M
17
57
Malformation
Tumor
Maxilla
Ear
Screws
Screws
Onlay
Onlay
Remodeling
Removal
529
THE JOURNAL OF CRANIOFACIAL SURGERY / VOLUME 16, NUMBER 4 July 2005
ever, exposure and infection are possible, especially
in grafts covered by thin and scarring soft tissues.
Consequently, some sites (i.e., nose, maxilla, and
ear) and diagnosis at admission (i.e., syndromic patients who have undergone several operations) are
risk factors that can affected the clinical outcome.
REFERENCES
Fig 5 Disease-specific survival rate calculated according
to type of fixation (x2 = 2.07, 1 df, P , 0.15, log rank test).
(1) Grafts fixated without screws (n = 59); (2) grafts fixated
with screws (n = 226).
Paresthesia is a minor complication reported in
literature. Ng et al11 describe a series of 30 patients
with orbital blowout fractures that were repaired using
porous polyethylene sheets. They reported one major
complication (a case of recurrent implant infection
leading to implant removal) and three minor postoperative complications (2 cases of postoperative infraorbital anesthesia and 1 case of a palpable titanium
screw). In our series, reconstruction of the orbital floor
is related to an high rate of paresthesia (11 of 49
cases), but this symptom was present preoperatively,
and it was associated with an orbital floor macrogap.
Consequently, it cannot be related to graft insertion.
Type of surgical insertion and type of fixation did
not reach statistical prognostic values. Although
a mixed system of insertion occurred in only 5.6%
of our series, these have a worst outcome (Fig 4). This
trend occurs because the higher the number of grafts
and surgical procedures, the higher is the risk of contamination. Because no difference was noted between
grafts fixed with or without screws, we can infer that
a correct implant integration can be obtained without
screws if the graft has a good primary stability.
In conclusion, porous polyethylene implants offer an excellent alternative to bone and cartilage in
reconstruction of many facial defects and deformities.
It is a biocompatible, strong substance that does not
resorb or degenerate. It is easy to work with, can be
fixated, and demonstrates long-term stability. How530
1. Wellisz T, Dougherty W, Gross J. Craniofacial applications for
the Medpor porous polyethylene flexblock implant. J Craniofac
Surg 1992;3:101–107
2. Couldwell WT, Chen TC, Weiss MH, et al. Cranioplasty with
the Medpor porous polyethylene flexblock implant. Technical
note. J Neurosurg 1994;81:483–486
3. Duman H, Deveci M, Uygur F, et al. Reconstruction of contour
and anterior wall defects of frontal bone with a porous polyethylene implant. J Craniomaxillofac Surg 1999;27:298–301
4. Park J, Guthikonda M. The Medpor sheet as a sellar buttress
after endonasal transsphenoidal surgery: technical note. Surg
Neurol 2004;61:488–492
5. Wellisz T. Clinical experience with the Medpor porous polyethylene implant. Aesthetic Plast Surg 1993;17:339–344
6. Frodel JL, Lee S. The use of high-density polyethylene implants
in facial deformities. Arch Otolaryngol Head Neck Surg 1998;
124:1219–1223
7. Sevin K, Askar I, Saray A, et al. Exposure of high-density porous polyethylene (Medpor) used for contour restoration and
treatment. Br J Oral Maxillofac Surg 2000;38:44–49
8. Wellisz T. Reconstruction of the burned external ear using
a Medpor porous polyethylene pivoting helix framework. Plast
Reconstr Surg 1993;91:811–818
9. Kim DY, Cho KS, Lee SY, et al. Surgical correction of cryptotia
using Medpor. Ann Plast Surg 1999;42:693–699
10. Romano JJ, Iliff NT, Manson PN. Use of Medpor porous polyethylene implants in 140 patients with facial fractures. J Craniofac Surg 1993;4:142–147
11. Ng SG, Madill SA, Inkster CF, et al. Medpor porous polyethylene
implants in orbital blowout fracture repair. Eye 2001;15:578–582
12. Karesh JW, Dresner SC. High-density porous polyethylene
(Medpor) as a successful anophthalmic socket implant. Ophthalmology 1994;101:1688–1695
13. Blaydon SM, Shepler TR, Neuhaus RW, et al. The porous polyethylene (Medpor) spherical orbital implant: a retrospective
study of 136 cases. Ophthal Plast Reconstr Surg 2003;19:364–371
14. Sagoo MS, Olver JM. Autogenous temporalis fascia patch graft
for porous polyethylene (Medpor) sphere orbital implant exposure. Br J Ophthalmol 2004;88:942–946
15. Wong JF, Soparkar CN, Patrinely JR. Correction of lower eyelid
retraction with high density porous polyethylene: The Medpor(R) Lower Eyelid Spacer. Orbit 2001;20:217–225
16. Tan J, Olver J, Wright M, et al. The use of porous polyethylene
(Medpor) lower eyelid spacers in lid heightening and stabilisation. Br J Ophthalmol 2004;88:1197–1200
17. Romo T, 3rd Sclafani AP, Sabini P. Use of porous high-density
polyethylene in revision rhinoplasty and in the platyrrhine
nose. Aesthet Plast Surg 1998;22:211–221
18. Niechajev I. Porous polyethylene implants for nasal reconstruction: clinical and histologic studies. Aesthet Plast Surg 1999;23:
395–402
19. Ozturk S, Sengezer M, Coskun U, et al. An unusual complication of a Medpor implant in nasal reconstruction: a case report.
Aesthet Plast Surg 2002;26:419–422
20. Dawson-Saunders B, Trapp RG. Basic Clinical Biostatistic. Norwalk: Appleton & Lange, 1994.
21. Cox DR, Oakes D. Analysis of Survival Data. New York: Chapman & Hall, 1984.