Survival following orbital exenteration at a tertiary brazilian hospital

Objective: to analyze the epidemiology, clinical features and survival rate of patients undergoing orbital exenteration (OE) in a tertiary referral hospital. Methods: we conducted a retrospective study of all patients undergoing OE at the Hospital das Clínicas, FMUSP between January 2007 and December 2012. We collected data records related to gender, age, origin, length of stay, duration of the disease, other treatments related to the disease, number of procedures outside of the face related to the disease, follow-up and histological diagnosis. Results: we treated 37 patients in the study period. The average survival in one year was 70%, in two years, 66.1%, and 58.3% in three years. There was no significant difference in the one-year survival related to histological diagnosis (p=0.15), days of hospitalization (p=0.17), gender (p=0.43), origin (p=0.78), disease duration (p=0.27) or the number of operations for the tumor (p=0.31). Mortality was higher in elderly patients (p=0.02). The average years of life lost was 33.9 in patients under 60 years, 14.7 in patients in the 61-80 years range and 11.3 in patients over 80 years. Conclusion: the present series of cases is significant in terms of prevalence of orbital exenteration; on the other hand, it shows one of the lowest survival rates in the literature. This suggests an urgent need for improved health care conditions to prevent deforming, radical resections.


INTRODUCTION INTRODUCTION INTRODUCTION INTRODUCTION INTRODUCTION
O rbit Exenteration (OE) is one of the most disfiguring procedures among ophthalmologic operations, and is characterized by the complete removal of the contents of the orbital cavity.According to the resection extent, it can be classified into: 1) total, if there is resection of the eyelids; 2) subtotal, when preserving the eyelids; or 3) extensive, when including removal of the bone surrounding walls 1 3 .
OE is the therapy of choice when other less radical methods do not result in better prognosis.It is usually indicated in oncologic resections for local control of malignant tumors.However, aggressive diseases or benign tumors that cause uncontrollable pain and structural and/ or extensive lesions also require it.Among the malignant lesions, Basal cell carcinoma (BCC) is the most common skin cancer (80-90%), followed by squamous cell carcinoma (SCC).Examples of non-malignant diseases include: neurofibromatosis, fibrous dysplasia, mucormycosis, sharply contracted anophthalmic cavity, recurrent meningioma and orbital myiasis 4,5 .
The aesthetic consequences have a strong psychological impact on the patient and require a multidisciplinary approach.Many patients are referred to psychological services after the operation or even refuse to undergo the surgical procedure.Constant vigilance, good doctor-patient relationship, early diagnosis and prompt treatment would provide better prognosis, especially in emerging countries 6,7 .
This retrospective study aims to analyze the epidemiology, clinical features and survival rate of patients undergoing orbital exenteration (OE) in a tertiary referral hospital.

METHODS METHODS METHODS METHODS
The research project was approved by the of the Hospital das Clínicas, University of São Paulo and we carried out a retrospective study of medical records and pathology reports of all patients who underwent orbital exenteration at the facility between January 2007 and December 2012.
We identified cases by the International Classification of Diseases (ICD-10).We requested the medical records and analyzed them manually.the following data were collected: gender, age, origin, days of hospitalization, time of disease, other operations/treatments performed related to the disease, number of procedures performed outside the area of the face related to the disease, follow-up, histologic diagnosis and recurrence of lesions.To analyze the survival rate, we contacted the patient's family members by telephone with the help of Social Service for identification and active search for the occurrence of death.
We analyzed the variables by the Kaplan-Meier method, and compared survival curves using the log-rank test, with the R software, version 3.1.1.We calculated the Years of Potential Life Lost (YPLL) by the method proposed by Romeder 8 , adjusted to the life expectancy of Brazilians in 2013 9 .The age of reference used was 78.6 for patients under 60 years of age, 83.7 for patients between 61 and 81 years, and 96.7 for patients over 80.

RESULTS RESULTS
We identified 39 patients, of whom two were excluded due to incorrect coding of the disease.

Demographic and clinical characteristics Demographic and clinical characteristics Demographic and clinical characteristics Demographic and clinical characteristics Demographic and clinical characteristics of patients of patients of patients of patients of patients
The study cohort included 17 men and 20 women, between 0 and 94 years of age (mean 62.2 years).São Paulo, capital, was the origin of 15 patients (40.5%), 13 (35.1%)were from towns in the interior of São Paulo and nine (24.4%) from other Brazilian regions.Thirty-three patients were white (89.2%), one was black (2.7%) and three brown (8.1%).
The average time of diagnosis was 43.4 months (range three months to 12 years), except for congenital cases.The days of hospitalization ranged from 0 to 62, average 14.Twelve patients (35.3%) were not subjected to any other surgical procedure related to the current injury, another 12 (35.3%)underwent one operation and 10 (29.4%) underwent more than one.Seventeen patients had additional treatment such as radiotherapy (ten patients -27%), chemotherapy (two patients -5.4%) and cryosurgery (three patients -8.1%).Most were not submitted to any other operation outside the face area (81.8%) and eight (21.6%) were previously treated at least once.

Survival Rate Survival Rate Survival Rate Survival Rate Survival Rate
We excluded congenital cases from the survival analysis.Two patients died during hospitalization.
At the time of the study, 15 patients had died, 15 were alive and six could not be contacted.The average survival rate at one year was 70% and this figure decreased to 66.1% and 58.3% in two and three years, respectively.Mean survival was 47.3 months.
The mortality rate was higher in older patients (p=0.02).There was no significant difference in one-year survival as for the histological diagnosis, if SCC (Figure 1), BCC or non-ECC/non-BCC (p=0.15),days of hospitalization (p=0.17),gender (p=0.43),origin (p=0.78),time of disease progression (p=0.27) or number of operations related to the tumor (p=0.31-Table 2).
The average age of death in the age group under 60 was 44.7 years; between 61 and 80 years, 69, and in patients aged over 80 years, 85.4.Considering the life expectancy of Brazil in 2013, the average years of life lost were, respectively, 33.9 years, 14.7 years and 11.3 years.The total YPLL was 191 years (Figure 2).As the hospital where the study was conducted is a tertiary center, it is expected that 59.9% of patients originate from other cities as well as from São Paulo.The geographical distance from the origin to the hospital also explains the choice for OE, as the imprecise diagnosis of other health services and lagged time to admission to the Figure 1 -Figure 1 -Figure 1 -Figure 1 -Figure 1 -Example of squamous cell carcinoma with orbital invasion.
tertiary hospital may have made OE the only possible procedure for the control of local disease.Among the patients cohort, three constituted nonmalignant cases.SCC and BCC together accounted for 70.2% of the histological diagnosis, which is consistent with other studies.BCC is the most common skin cancer in the periorbital area, but SCC spreads more easily and requires a quick management to prevent disease progression 2,10,12,14,15 .Our findings are similar to current literature, insofar as BCC represented 27% of the OE cases, while SCC accounted for 43.2%.Although SCC is more aggressive than BCC, the difference in survival at one year was not statistically significant between histopathologic diagnoses (p=0.15).The difference was evident only among the first 30 months or so.Some studies, however, had higher mortality after SCC Rev. Col. Bras.Cir.2016; 43(1): 042-047 than BCC [16][17][18] .Additional treatments, such as Mohs micrographic surgery, may have been beneficial in the management of some SCC cases 19,20 .
The average mortality rate after OE also differs from the literature, since our series showed lower survival.Rahman et al. reported a survival rate of 93% in one year 10 ;  Younger patients had on average 33.9 years of life lost as a result of diseases that lead to OE, and older patients lost more than ten years.Not only the aggressiveness of the disease, but also the lack of information, difficulty in access to health care and delay in correct diagnosis justify the current low survival rate 6,21 .Studies suggest differences in post-SCC mortality between developed and developing countries 22 .Advanced age may act as a confounding variable because, generally, it is related to comorbidities and other causes of death unrelated to the tumor.However, the predominance of advanced malignant disease is already an indicator of difficulty in access to adequate medical services for immediate treatment, which could improve survival even in the older age group.
In conclusion, this case series is significant in terms of prevalence of Orbit Exenteration; On the other hand, it displayed one of the lowest survival rates in the literature.This suggests an urgent need for improved health care conditions to prevent deforming, radical resections.
DISCUSSIONDISCUSSION DISCUSSION DISCUSSION DISCUSSION Orbital exenteration is not a common procedure and is usually done in tertiary referral centers.Our case series presented one of the largest series per year (37 patients in six years).Rahman et al. reported 64 cases in a period of 13 years 10 ; Mohr and Esser had 77 in 20 years 11 ; Bartley et al. described 102 in 20 years 12 ; and Maheshwari et al. published 15 in 10 years 13 .

Table 2 -Table 2 -Table 2 -Table 2 -Table 2 -
Comparison of age, gender, days of hospitalization, origin, time of disease, number of operations and histological diagnosis with survival rate.