DERMATOLOGIC SURGERY

Life expectancy after Mohs micrographic surgery inpatients aged 90 years and older

Amy Delaney, MD,a Ikue Shimizu, MD,b Leonard H. Goldberg, MD,c and Deborah F. MacFarlane, MD, MPHb

Danville, Pennsylvania, and Houston, Texas

From

D

A

H

Fund

Conf

296

Background: The population of people aged 90 years and older is expected to more than triple by 2050.The incidence of skin cancers is increasing.

Objective: We sought to determine whether treatment of patients aged 90 years and older with skin cancerby Mohs micrographic surgery (MMS) changed their survival.

Methods: A group of 214 patients aged 90 years and older who underwent MMS from July 1997 to May2006 was identified. Patient gender, age, tumor type, size, site, defect size, number of MMS stages, andsurgical repair were recorded. Comorbid medical conditions were assessed using the Charlson index.Actual survival was compared with expected length of survival using life tables. Data were analyzed by theKaplan-Meier method with log rank significance tests.

Results: Average patient age was 92.3 years. All patients tolerated the procedures well with no deathswithin 1 month after surgery. Median survival after surgery was 36.9 months. Tumor characteristics, defectsize, number of surgical stages, and closure type did not affect survival. There was no significant differencein survival based on comorbidities according to Charlson scores. Instantaneous mortality hazard washighest 2 to 3 years after surgery.

Limitations: Specific causes of death were not accessible.

Conclusion: This growing section of the population may safely undergo MMS. ( J Am Acad Dermatol2013;68:296-300.)

Key words: basal cell carcinoma; Mohs micrographic surgery; nonagenarian; perioperative mortality;perioperative safety; squamous cell carcinoma.

ccording to the latest figures from the US

Abbreviations used:

CC: Charlson classCS: Charlson scoreMMS: Mohs micrographic surgeryNMSC: nonmelanoma skin cancer

A Census Bureau, the population of peopleaged 90 years or older is predicted to in-

crease from 1.45 million in 20091 to 8.7 million by theyear 2050, when the US population should total 439million.2 Given the increasing incidence of non-melanoma skin cancers (NMSCs) in the UnitedStates, physicians will consequently be faced withan increasing number of very elderly patients pre-senting with skin cancers.

This study was therefore designed to assess thesafety of Mohs micrographic surgery (MMS) in no-nagenarians with NMSCs, and to investigate if thereare any factors that can be used to predict shorter

the Department of Dermatology, Geisinger Medical Center,

anvillea; Department of Dermatology, University of Texas, MD

nderson Cancer Centerb; and DermSurgery Associates,

ouston.c

ing sources: None.

licts of interest: None declared.

survival in this subgroup. We intend to evaluate thepotential effect of medical comorbidities on survivalby using the Charlson index, a validated measure-ment tool that has been used to predict survivalbased on the presence of comorbidities such as heartdisease or diabetes.3,4

Reprint requests: Deborah F. MacFarlane, MD, MPH, MD Anderson

Cancer Center, 1400 Pressler St, Unit 1452, Houston, TX 77030.

E-mail: dmacfarl@mdanderson.org.

Published online November 29, 2012.

0190-9622/$36.00

� 2012 by the American Academy of Dermatology, Inc.

http://dx.doi.org/10.1016/j.jaad.2012.10.016

J AM ACAD DERMATOL

VOLUME 68, NUMBER 2Delaney et al 297

METHODSInstitutional review board approval for a retro-

spective study was obtained. Records of all patientswho underwent MMS by the same Mohs surgeon at 2practice sites (an established private MMS practiceand a hospital-based outpatient Mohs clinic) fromJuly 1, 1997, to May 31, 2006, were screened by age.

CAPSULE SUMMARY

d Although cutaneous surgery isassociated with low morbidity andmortality, the impact of complex Mohsmicrographic surgery on nonagenariansis unknown.

d This study demonstrates that complexityof surgery (number of Mohs stages,repair) or medical comorbidities do notaffect perioperative mortality.

d Nonagenarians can safely undergo Mohsmicrographic surgery and haveappropriate repairs without beingrelegated to less effective options forfear of prolonged or complex cases.

There were 214 patients whowere aged 90 years or olderat the time of initial surgery;123 patients were operatedon in the private dermatol-ogy clinic and 91 patientsunderwent MMS in thehospital-based outpatientclinic. Data regarding age attime of surgery, gender, tu-mor type, tumor site, preop-erative and postoperativesize of the tumor and defect,number of Mohs stagesneeded for clearance, andtype of closure were ex-tracted. Patient race was notrecorded, although the ma-jority of patients were knownto be Caucasian. Surgery was

performed under local anesthesia (buffered 0.05%lidocaine with epinephrine 1:200,000, used by thesenior authors for the past 2 decades on all patients,with excellent hemostasis and very few side effects).In patients who reported an allergy to lidocaine orepinephrine, 0.25% bupivacaine was used. Any sur-gical complications were noted.

The Charlson index is a weighted index used toquantify the effect of comorbid medical conditions.This index assigns a value for different comorbidconditions that is directly proportional to the nega-tive impact upon life expectancy. The total value ofall comorbidities for a given patient is the Charlsonscore (CS). A CS of 3 or greater predicts shortersurvival, although it remains difficult to predict thesurvival of individual patients with scores greaterthan or equal to 3.4 The CS for each patient wascalculated, and each patient was then grouped into aCharlson class (CC) by CS: CC0 = CS0, CC1 = CS1 toCS2, CC2 = CS3 to CS4, and CC3 = CS5 or greater.

In addition, patients’ social security numbers wereused to search for the date of death for deceasedpatients in the National Death Index via Ancestry.com. This portal is widely used for conductinggenealogic research and surveys that do not containmany patients, as direct queries of the National DeathIndex have impractical submission requirements forresearch involving fewer than a thousand records.

The lag time varies, but is typically within the range of3 months. Survival times were calculated in monthsfrom the timeof initial surgery until deathor the endoffollow-up in September 2006. The length of survivalwas comparedwith the expected lengthof survival forthe US population based on life tables obtained fromthe National Center for Health Statistics. Data were

analyzed by the KaplanMeiermethod with log rank signifi-cance tests.5 For those pa-tients who had multiplesurgeries after the age of 90years, the first surgery wastaken as the index case.

RESULTSAverage patient age at

time of initial surgery was92.3 years (range 90-101years). There were 107 menand 107 women. All tumorswere primary, with a total of148 basal cell carcinomasand 66 squamous cell carci-nomas (ratio 2.24:1). Basalcell carcinoma was morecommon among women

than men, and occurred most commonly on thenose, whereas squamous cell carcinoma was morecommon in men and was most frequent on theextremities. No regional metastatic disease was iden-tified. All patients tolerated their procedures withoutcomplications. The first death occurred at 33 daysafter surgery, but follow-up notes indicated nocomplications related to MMS. Median survival was36.9 months after surgery. The longest survivor wasstill alive 108 months after surgery.

Tumor type, tumor size, location, or defect sizedid not affect survival (Table I). The number of stagesof surgery performed ranged from 1 to 8; approxi-mately half of the patients had only 1 stage (n = 105)and a third underwent 2 stages (n = 70). There wasno difference in survival between the patients whounderwent 1 or 2 stages and those who had 3 ormore stages (Fig 1) (P\ .41). The patient with theshortest survival time of 33 days underwent 6 stagesof surgery, whereas the patient who had 8 stages ofsurgery was still alive at the end of follow-up 69months after surgery. A total of 111 linear closures, 53flaps, and 19 grafts were performed. Closure typewas unavailable for 3 patients. No difference insurvival was noted with respect to closure type(P\ .4) (Fig 2).

A total of 102 patients were grouped into CC0(CS0), 74 were in CC1 (CS1-CS2), 21 were in CC2

Table I. Effects of tumor characteristics uponsurvival

Variable Median survival, mo P value

Basal cell carcinoma 37.9 .69Squamous cell carcinoma 29.5Tumor size, cm .25\0.5 32.91.0 41.91.5 39.4[1.5 25.3

Tumor location .53Extremity 40.4Trunk 42.5Head 33.2

Defect size, cm .550.5 62.11.0 33.21.5 33.1

No. of stages .511 37.92 32.9$ 3 43.2

Fig 1. Effect of number of stages of Mohs upon patientsurvival.

Fig 3. Effect of comorbidities upon patient survival.

Fig 2. Effect of closure type upon patient survival.

J AM ACAD DERMATOL

FEBRUARY 2013298 Delaney et al

(CS3-CS4), and 17 were in CC3 ($CS5). Fig 3 showsthat there is no significant difference in survivalbased on CC. There were patients with multiple orserious comorbid conditions (higher CS) who sur-vived many years, and patients with 0 or 1 comorbidmedical conditions who survived only a fewmonths.

The instantaneous mortality hazard (Fig 4) wasthe highest 2 to 3 years after surgery. There was noincreased perioperative mortality. A statistical com-parison of the surgical cohort with the age-adjustedUS population showed no difference in the expectedsurvival. The age of the patients at the time of surgerywas unrelated to survival outcome. The only

significant risk factor identified was gender, withwomen experiencing a significant survival advan-tage over men (P\ .02).

DISCUSSIONCutaneous surgeryunder local anesthesia has been

consistently associated with low rates of morbidityand mortality.6,7 Two previous studies have demon-strated that survival for nonagenarian patients under-going MMS is similar to survival for nonagenarians inthe general populationnotundergoing surgery. In ouroriginal study, 115 patients aged 90 years or olderunderwent MMS from January 1988 to August 1996 ina hospital-based setting.8 Only 1 complication oc-curred when a 93-year-old woman with a pacemakerhad chestpainduring surgery.Her chest pain resolved

Fig 4. Mortality hazard after surgery.

J AM ACAD DERMATOL

VOLUME 68, NUMBER 2Delaney et al 299

and her operative site was closed the following daywithout any complications. This study showed thatMMS was well tolerated and could be safely per-formed in an elderly population without increasingthe risk of perioperative morbidity or mortality.

In a subsequent study, Charles et al4 reviewed thecharts of 99 patients aged 90 years and older who hadMMS for NMSCs from January 1985 to June 1994 in ahospital-based setting to determine prognostic fac-tors for survival. The authors converted comorbid-ities using the Charlson index into CS to determinewhether comorbid conditions affected patients’ sur-vival. Charles et al4 reported that patients with highCS (multiple comorbidities) had a shorter survivalthan those with low scores, and concluded that theCharlson index is a valid prognostic factor forpredicting life expectancy in the very elderly withNMSCs who undergo MMS.

The current study assesses the impact of the dura-tion of surgery as measured by the number of surgicalstages and the type of closure on survival in thesepatients. Patients who underwent fewer stages ofsurgery did not survive any longer than those whounderwent 3 or more stages. Interestingly, patientswhose defects were allowed to heal by second inten-tion did not survive any longer than those patientswhose defects were repaired with flaps or grafts. Itappears that patients older than 90 years can toleratelonger procedures, including multiple stages andcomplex repair; there is no need to assume that theelderly patient is best repaired with the simplestsurgical option. This supports an earlier observationby Shumick et al9 that forehead flaps and cartilagegrafts can be performed without significant morbidityin appropriately chosenpatients ages 80years orolder.

The CS did not reliably predict survival in ourstudy population; patients with multiple comorbid-ities survived as long as patients with only a few or

no medical comorbidities. This is not surprisinggiven that individuals with many serious comorbid-ities are likely to die before age 90 years. In addition,one is unlikely to operate on obviously clinicallyunstable or very ill patients; treatment of skin cancerin such a patient would not be a priority. Basically,the group of nonagenarians for whom skin cancertreatment would be considered is a self-selectedgroup of relatively robust individuals, and thus thenumber of comorbidities as reflected by CS wouldnot accurately predict survival.

It is noteworthy that there were no complicationsin our patient population. In particular, there wereno complications in those patients who had surgeryperformed at the private practice site, which sup-ports the idea that MMS in nonagenarians do notneed to be performed in a hospital-based setting. Inaddition, the chance of dying after surgery, asassessed by the instantaneous mortality hazard,was highest at 2 to 3 years after surgery. If therewere a causal relationship between MMS and de-creased patient survival, this measure would havepeaked in the perioperative period. The causes ofdeath were not accessible, but none of the patientswho died within the first 3 months after MMS hadcomplications documented during the postoperativeperiod. This supports the statement that no deathswere attributable to MMS. Finally, there was asignificant survival advantage of women over menin our study, but this is to be expected, as the averagelife expectancy in the United States for the veryelderly is greater for women than for men.

Some precautions are appropriate when perform-ing MMS on nonagenarians. Any precautions appro-priate for the elderly patient are also applicable to thenonagenarian. These include having a family mem-ber or companion present, ensuring comfort andtemperature control (blankets, pillows), access toadequate nutrition, and being aware of nonverbalcues of distress. Extra caution should also be exer-cised with regard to current medications, paincontrol including lidocaine dosage, hemostasis, per-ioperative transportation issues, wound care ability,and any underlying health issues. Primary carephysicians should be consulted as needed. Inessence, awareness of the overall health of thepatient is important. Age is simply one factor ofmany that may affect health status.

In summary, this article analyzes the prognosticfactors for survival in patients aged 90 years andolder based on age, gender, comorbidities, numberof surgical stages, type, size and location of tumor,and type of repair. Except for the expected survivaladvantage of women over men in this age group, nohealth status or procedure-related effects on

J AM ACAD DERMATOL

FEBRUARY 2013300 Delaney et al

mortality were identified. This growing section of thepopulation can safely undergo MMS and should notbe relegated to other less effective treatmentsthrough fear of affecting their survival.

REFERENCES

1. US Census Bureau, Population Division. Annual estimates of the

resident population by sex and five-year age groups for the

United States: April 1, 2000 to July 1, 2009. Available from: URL:

http://www.census.gov/popest/data/national/asrh/2009/tables/

NC-EST2009-01.xls. Accessed March 2, 2012.

2. US Census Bureau, Population Division. Projections of the popu-

lationby ageand sex for theUnitedStates: 2010 to2050.Available

from: URL:http://www.census.gov/population/www/projections/

files/nation/summary/np2008-t12.xls. Accessed March 2, 2012.

3. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method

of classifying prognostic comorbidity in longitudinal studies:

development and validation. J Chronic Dis 1987;40:373-83.

4. Charles AJ, Otley CC, Pond GR. Prognostic factors for life

expectancy in nonagenarians with nonmelanoma skin cancer:

implications for selecting surgical candidates. J Am Acad

Dermatol 2002;47:419-22.

5. Kalbfleisch JD, Prentice RL. The statistical analysis of failure time

data. New York: John Wiley and Sons; 1980.

6. Kimyai-Asadi A, Goldberg LH, Peterson SR, Silapint S, Jih MH.

The incidence of major complications from Mohs micrographic

surgery performed in office-based and hospital-based settings.

J Am Acad Dermatol 2005;53:628-34.

7. Amici JM, Roges A, Lasheras A, Gachie JP, Guillot P, Beylot C, et al.

A prospective study of the incidence of complications associated

with dermatologic surgery. Br J Dermatol 2005;153:967-71.

8. MacFarlane D, Goldberg L, Pustelny B. An assessment of the

suitability of Mohs micrographic surgery in patients aged 90

years and older. Dermatol Surg 1997;23:389-93.

9. Shumick KA, Campbell A, Becker F. Nasal reconstruction in the

elderly patient: the case of not letting age determine method.

Arch Facial Plast Surg 1999;1:297-301.