Ahos Pice Home Care Sys Tem For Ad Vanced Can Cer
Chi nese Med i cal Jour nal (Tai pei) 2002;65:331-335
Orig i nal
A Pro spec tive Study of Pain Treat ment for
Pa tients with Ad vanced Can cer Who
Re ceive Hos pice Home Care
Wei-Shou Hwang1
Back ground. Pain is of ten in ad e quately treated in pa tients with can cer.
Yu-Fang Tsai2
Pre vi ous stud ies have dem on strated the ef fec tive ness of the World Health
Hsien-Chen Chang2
Or ga ni za tion (WHO) an al ge sic lad der in can cer pain man age ment.
I-Ping Liu2
Methods. A to tal of 131 con sec u tive pa tients with ad vanced can cer re -
Chien-Tai Huang1
ferred to a hos pice home care pro gram were en rolled over one year pe riod
from Jan. 1 to Dec. 31, 2000. We as sessed the ad e quacy of pre scribed an al -
1Di vi sion of He ma tol ogy/On col ogy,
ge sic drugs us ing guide lines de vel oped by the WHO. Age, gen der, East ern
De part ment of In ter nal Med i cine; and
Co op er a tive On col ogy Group per for mance sta tus, pain mech a nism at re -
2De part ment of Nursing, Chi-Mei
fer ral, pain and symp tom in ten sity, and doses and days of drug ad min is tra -
Foun da tion Hos pi tal, Tainan, Tai wan,
tion dur ing the course of treat ment were re corded at reg u lar in ter vals.
R.O.C.
Re sults. Eighty-two per cent of the pa tients (107 of 131) had pain symtoms
at re fer ral. Forty-seven pa tients were ex cluded from this study due to in ad -
Key Words
e quate fol low-up times or in abil ity to ex press the pain in ten sity. Sixty pa -
can cer pain;
tients who had mea sur able pain in ten sity re quir ing an al ge sic ther apy were
hos pice home care;
fol lowed up un til death for a mean du ra tion of 65 days. At re fer ral, 46% of
World Health Or ga ni za tion guide lines
the pa tients (28 of 60) re ceived in ad e quate treat ment. In the last week of
life, 2%, 26% and 70% of pa tients were tak ing non-opioid drugs, mod er -
ate opioids and strong opioids, re spec tively. A sig nif i cant im prove ment in
pain and symp tom in ten sity was achieved af ter re fer ral. A mi nor ity of the
pa tients (10%) had in ad e quate pain con trol in the last week of life.
Con clu sions. This study dem on strates that a man aged hos pice home care
sys tem en ables pa tients to re ceive ad e quate pain treat ment, ac cord ing to
WHO guide lines. [Chin Med J (Tai pei) 2002;65:331-335]
A hos pice home care sys tem for ad vanced can cer ad vanced can cer pa tients. Ex perts of the World Health
pa tients was started in 1983 and is greatly ex -
Or ga ni za tion (WHO) have de vel oped guide lines that
pend ing in Tai wan in re cent years. A large pro por tion
aim to im prove the man age ment of can cer pain.7 The
of pa tients with met a static can cer have pain long be -
fea si bil ity and ef fi cacy of the WHO an al ge sic lad der
fore the ter mi nal stage of their ill ness.1-5 Many pa tients
has been re ported in dif fer ent stud ies. This pro spec -
with can cer have con sid er able pain and re ceive in ad e -
tive study in cluded pa tients with ad vanced can cer who
quate an al ge sia.5,6 Con trol of symp toms, in clud ing
re ceived fol low-up at home to de ter mine the ef fec tive -
pain, is the fun da men tal aim of the pal lia tive care of
ness, safety and fea si bil ity of a pain man age ment
Re ceived: September 11, 2001. Ac cepted: March 18, 2002.
Cor re spon dence to: Wei-Shou Hwang, MD, Di vi sion of He ma tol ogy/On col ogy, De part ment of In ter nal Med i cine, Chi-Mei Foun da tion
Hos pi tal, 901, Chung-Hwa Road, Tainan 710, Tai wan. Fax: +886-282-8928; E-mail: 7300@mail.chimei.org.tw
332 Wei-Shou Hwang, et al.
Chi nese Med i cal Jour nal (Tai pei) Vol. 65 No. 7
based on the WHO method.
rized pain in ten sity be tween two time points. A level
of p < 0.05 was con sid ered sig nif i cant in all tests, ex -
cept mul ti ple com par i sons fol low ing Fried man
Methods
two-way ANOVA test.
Be tween Jan u ary 1 and De cem ber 31, 2000, one
hun dred and thirty-one con sec u tive pa tients re ferred
Re sults
to hos pice home care pro gram at Chi-Mei Fundation
Hos pi tal were stud ied. The hos pice home care team
One-hundred and seven pa tients (82%) had pain at
con sisted of two oncologists, two nurses, a so cial
re fer ral. Forty-seven pa tients were ex cluded from the
worker and a psy chi at ric doc tor. The home vis its were
study be cause they had too short a pe riod of fol low-up
per formed twice a week by the nurses and weekly by
or were in abil ity to ex press the pain in ten sity. Twenty-
the doc tor, de pend ing on the pa tient’s con di tion. A
four pa tients did not need any an al ge sic treat ment un -
con tin u ous on-call sys tem was set up. Data were col -
til death. The re main ing 60 pa tients pre sented with
lected and re corded in spe cial files in di cat ing symp -
pain re quir ing an al ge sic ther apy and were fol lowed
toms, drug doses and gen eral in for ma tion about the
un til death and also in cluded in anal y sis (Ta ble 1). The
pa tients. An al ge sic was ad min is tered ac cord ing the
mean du ra tion of home care was 65 days. Twenty-one
WHO an al ge sic lad der. Pain in ten sity was mea sured
us ing the pa tient’s self re port on a nu meric scale from
Table 1. Patient characteristics
0 to 10, which was mea sured by Vi sual An a logue
Total number of patients
131
Scale (VAS). Step or opioid dos age were changed to
Patients with pain requiring the use of analgesics
107
im prove pain re lief when the pain in ten sity was con -
Dropouts
47
sid ered un ac cept able by pa tients. The use of other
Number of patients available for study
60
Gender (M/F)
24/36
drugs were al lowed, in clud ing the ones gen er ally ad -
Number of pain sites (%)
min is tered in pal lia tive care to con trol symp toms due
1
70
to ill ness or treat ment. Emo tional sup port was of fered
2
23.4
in weekly dis cus sion groups to help al le vi ate the psy -
≥ 3
6.6
cho logic prob lems of pa tients and their fam i lies. At
Site of primary tumor (%)
the ini tial vis its, age, gen der and East ern Co op er a tive
Gastrointestinal
21.7
On col ogy Group (ECOG) per for mance sta tus were
Lung
18.3
re corded. The data of pain syn dromes, ECOG scales,
Uro-gynaecological
16.7
pain in ten sity, symp toms as so ci ated with opioid ther -
Liver
8.3
apy, and the doses of drugs dur ing the course of treat -
Head and neck
6.7
Breast
6.7
ment were re corded at re fer ral (T0), af ter one week
Pancreas
3.3
(T1) and dur ing the last week of life (T2). The dis tri bu -
Others
18.3
tion of VAS was ex am ined for T , T , T and ex -
0
1
2
pressed with mean and me dian. The VAS’ dif fer ences
Table 2. Types of pain
from the pairs (T and T , T and T , T and T ) were
0
1
0
2
1
2
tested by means of Fried man two-way ANOVA test.
Somatic (%)
Visceral (%)
Neuropathic (%)
Wilcoxon Signed-Rank Test with a Bonferroni cor -
Pure
3 (5)
25 (41.7)
13 (21.7)
rected al pha level (0.05/3) was used as a post hoc mul -
SV
9 (15)
9 (15)
SN
6 (10)
6 (10)
ti ple com par i son method. We also dichotomized VAS
VN
6 (10)
6 (10)
with a cut-off point of four: more pain ful and less
SVN
2 (3)
2 (3)
2 (3)
pain ful. McNemar’s test was per formed once at each
Total
20 (38)
42 (70)
27 (45)
of the three time points to test the change of cat e go -
N = Neuropathic; S = Somatic; V = Visceral.
July 2002
Pain Treat ment for Hos pice Can cer Pa tients 333
Table 3. Number of patients on different steps at referral
pa tients were with step 1 and 2 (5%), and 1 pa tient
(T ), after 1 week of treatment (T ), and in the
0
1
was with step 1, 2 and 3 (1%). The move ment of pa -
last week of life (T2)
tients through the dif fer ent steps at re fer ral, af ter 1
n
Step 1
Step 2
Step 3
week and in the last week of life is pre sented in Ta ble 3.
T
Pain in ten sity and the num ber of pa tients with un -
0
60
5
18
37
T1
60
3
20
37
con trolled pain are shown in Ta ble 4. Sig nif i cant dif -
T2
60
1
16
43
fer ences were ob served in pain in ten sity at T1 and T2
when com pared with T0 (p < 0.005). The mean vi sual
Table 4. Pain intensity and performance status
an a logue scale (VAS) at re fer ral was 4.58, while VAS
at T and T were 2.80 and 2.02 re spec tively. The in di -
1
2
T0
T1
T2
ca tions for us ing al ter na tive routes to the oral one at
VASa,b
4.58-4.00
2.80-2.00c
2.02-2.00c,d
the study in ter vals are shown in Ta ble 5. The sub cu ta -
VAS ≤ 4, > 4e, f
32-28c
51-9c
54-6c
Number of patients
60
60
60
ne ous route and trans derm al fentanyl were the most
ECOG scale
3
3
4
fre quently used, mainly in the last days of life. The
a
mean daily dose of oral mor phine was 65 mg. The
Wilcoxon signed rank test as a post hoc comparison
followed with Friedman two way ANOVA; bVAS was
most com mon side ef fects of opioid an al ge sics were
expressed as mean-median; cp < 0.001 when compared with
con sti pa tion (20%), nau sea and vom it ing (5.1%), but
T0 ; dp = 0.001 when compared with T1; eMcNemar’s test;
the in ten sity level of the side ef fects was ac cept able in
fthe digits express the number of patients in each group.
most pa tients. The most fre quently used drugs for
ECOG = Eastern Cooperative Oncology Group; VAS =
adjuvant treat ment in cluded coanalgesics of Non-
Visual Analogue Scale.
steroid anti-inflammatory drugs, ste roids, lax a tives
pa tients were ad mit ted to the hos pi tal. The mean
and antiemetics.
length of in-patient hos pi tal iza tion was 11.5 days.
Hos pi tal iza tion was due to un con trolled pain in 13%
of these 21 pa tients.
Dis cus sion
The most com mon sites of pri mary tu mors were
the gas tro in tes ti nal tract and lung. Sev enty per cent of
Dif fer ent stud ies have dem on strated the ef fec tive -
pa tients had only one pain site (Ta ble 1). Pain mech a -
ness of the WHO an al ge sic lad der in can cer pain man -
nisms are listed in Ta ble 2. There were 20 so matic
age ment.8-10 Dr. Mercadante car ried out a study of an -
pain syn dromes (38%), 42 vis ceral pain (70%) and 27
al ge sic treat ment based on the WHO method for ad -
neuropathic pain (45%). In most in stance they were
vanced can cer pa tients who re ceived fol low-up at
com bined. The mean du ra tion pe riod of step 1, 2 and 3
home un til death.11 In that study, most pa tients at re -
were 5 (range, 12-32), 20 (range, 10-208), and 55
fer ral and af ter 1 week of treat ment had the per for -
(range, 3-203) days, re spec tively. Only 5 pa tients re -
mance sta tus of 2 (ECOG Scale). Our study in cluded
ceived nonopiod analgesie an al ge sics alone (8%), 3
ad vanced can cer pa tients with per for mance sta tus
Table 5. Use of alternative route before admission (T0), after 1 week (T1), and in the last week of life (T2)
T0
T1
T2
S.C.
TTS - Fentanyl
S.C.
TTS - Fentanyl
S.C.
TTS - Fentanyl
PCA
Alternative routes
1
4
3
5
11
8
6
Due to nausea,
-
1
-
1
-
-
-
Due to confusion or drowsiness
-
-
-
1
-
1
-
Due to inability to oral intake
-
2
-
3
2
5
4
Due to uncontrolled pain
-
1
-
-
-
2
2
PCA = Patient Controlled Anagesia; SC = Subcutaneous; TTS = Transdermal Therapeutic System.
334 Wei-Shou Hwang, et al.
Chi nese Med i cal Jour nal (Tai pei) Vol. 65 No. 7
greater than 3 on the ECOG scale who were taken care
are dif fi cult to eval u ate be cause they in volve dif fer ent
of in the hos pice home care sys tem. The an al ge sic
drugs and dos ages. Lax a tives and antiemetics were
treatmant was based on the WHO method. The per -
com monly used to limit opioid-related side ef fects.
cent age of our pa tients with pain re quir ing an al ge sic
Ste roids were used for dif fer ent in di ca tions.
was 82%, which was sim i lar to other re ports.13 A to tal
Our re sults sug gest that more than 100 ad vanced
of 98% of pa tients with pain re ceived opioids be fore
can cer pa tients per year are able to ben e fit from a pal -
death. Of those, 29% re ceived a weak opioid (step 2)
lia tive home care for vary ing pe ri ods of time (2
with a mean du ra tion of 20 days. In a large study, 24%
months on av er age) at our hos pi tal. This study dem on -
of 871 pa tients were treated with step 2 pro to col with
strates that a man aged hos pice home care sys tem en -
a mean du ra tion of 28 days.8 The mean dos age of oral
ables pa tients to re ceive ad e quate pain treat ment, ac -
mor phine in our study was 65 mg, which was sim i lar
cord ing to WHO guide lines.
to Mercadante’s re port13 and lower then than other
groups.14 A wide vari a tion in the mean daily dose of
mophine has been re ported by dif fer ent cen ters. The
Achnowledgements
mean max i mum dose was 60-90 mg.15-17 To de fine
pain se ver ity, the nu meric rat ing scale has proven
We thank the grand grant (No.89001) sup port
pop u lar for ad ap ta tion to home care. Three lev els of
from Hos pice Foun da tion of Tai wan.
pain se ver ity interferring with the pa tient’s func tion
have been es tab lished, in which rat ings of 1-4 cor re -
spond to mild pain, 5-6 to mod er ate pain and 7-10 to
Ref er ences
se vere pain. Be fore referal to hos pice home care pro -
gram, our pa tients were taken care in dif fer ent de part -
1. Daut RL, Cleeland CS. The prev a lence and se ver ity of pain
ments. The can cer pain was of ten poorly man aged.
in can cer. Can cer 1982;50:1913-8.
2. Green wald HP, Bonica JJ, Bergner M. The prev a lence of
De spite 46% of our pa tients be ing with sig nif i cant
pain in four can cers. Can cer 1987;60:2563-9.
pain at re fer ral, only a mi nor ity of pa tients had in suf fi -
3. Peteer J, Tay V, Co hen G, MacIn tyre J. Mac In tyre J. Pain
cient pain re lief af ter 1 week of treat ment and in the
char ac ter is tics and treat ment in an out pa tient pop u la tion.
last week of life (15% and 10%, re spec tively). Sat is -
Can cer 1986;57:1256-65.
fac tory pain con trol was pos si ble in 90% of our cases.
4. Portenoy RK. Can cer pain: epidermiology and syndrone syn -
dromes. Can cer 1989:2298-907; 1989;63:2298-307.
Non-oral route opioid ad min is tra tion clearly in -
5. Cleeland CS, Gonin R, Hat field AK, Edmonson JH, Blum
creased with the pro gres sion of the un der ly ing dis -
RH, Stew art JA, et al. Pain and its treat ment in out pa tients
ease, mainly in the last week of life, due to neu ro logic
with met a static can cer. N Engl J Med 1994;330:592-6.
de range ments or the pres ence of nau sea, vom it ing or
6. Von Roenn JH, Cleeland CS, Gonin R, Hatfild AK, Pandya
un con trolled pain. Sim i lar find ings were re ported for
KJ. Phy si cian at ti tudes and prac tice in can cer man age ment: a
sur vey from the East ern Co op er a tive On col ogy Group. Ann
dy ing pa tients;12 45% of our pa tients re quired
In tern Med 1993;119:121-6.
parenteral treat ment. In re cent years, the de vel op ment
7. World Health Organigation. Can cer Pain Re lief. Geneva:
of trans derm al opioid an al ge sics has been a ma jor ad -
World Health Or ga ni za tion, 1986.
vance in the better com pli ance and in the abil ity of pa -
8. Ventafridda V, Tambuvini M, Caraceni A, De Conno F, Naldi
tients to sleep through the night for sev eral un in ter -
F. A val i da tion study of the WHO method for can cer pain re -
rupted hours.18 Sev eral stud ies re ported that trans -
lief. Can cer 1987;59:850-6.
9. Walker VA, Hoskin PJ, Hank GW, White ID. Eval u a tion of
derm al fentanyl pro vided pain re lief and was as so ci -
WHO an al ge sic guide lines for can cer pain in a hos pi -
ated with less con sti pa tion and se da tion than mor -
tal-based pal lia tive care unit. J Pain Symp tom Man age
phine.19,20 In 13% of our pa tients, the an al ge sic treat -
1998;3:145-9.
ment was ad min is tered with trans derm al fentanyl in
10. Zech DFJ, ground Ground S, Lynch J, Hertel D, Lehman KA.
the last weeks. In other stud ies, dif fer ent rates of
Val i da tion of World Health Or ga ni za tion guide lines for can -
cer pain re lief: a 10-year pro spec tive study. Pain 1995;
coanalgesic use have been re ported. The treat ments
63:65-76.
July 2002
Pain Treat ment for Hos pice Can cer Pa tients 335
11. Mercadante S. Pain treat ment and out comes for pa tients with
17. Boisvert M, Co hen SR. Opioid use in ad vanced ma lig nant
ad vanced can cer who re ceive fol low-up care at home. Can -
dis ease: Why do dif fer ent cen ters use vastly dif fer ent doses?
cer 1999;85:1849-58.
A plea for stan dard ized re port ing. J Pain Symp tom Man age
12. Grond S, Zech D, Schug SA, Lynch J, Lehmann KA. Val i da -
1995;10:632-8.
tion of World Health Or ga ni za tion guide lines for can cer pain
18. Serlin RC, Mendoza TR, Nakamura Y, Ed wards KR,
re lief dur ing the last days and hours of life. J Pain Symp tom
Cleeland CS. When is can cer pain mild, mod er ate or se vere?
Man age 1991;6:412-22.
Grading pain se ver ity by its in ter fer ence with func tion. Pain
13. Mercadante S, Genovese G, Kargar JA, Maddaloni S,
1995;61:277-84.
Roccella S, Salvaggio L, et al. Home pal lia tive care: re sults
19. Payne R, Chan dler S, Einhaus M. Guide lines for the clin i cal
in 1991 ver sus 1988. J Pain Symp tom Man age 1992;7:414-8.
use of trans derm al fentanyl. Anticancer Drug 1995;6:50-3.
14. Mount B. Med i cal ap pli ca tions of her oin. Can Med Assoc J
20. Ahmedzai S, Brooks D. Trans derm al fentanyl ver sus sus -
1979;120:405-7.
tained-release oral mor phine in can cer pain: pref er ence, ef fi -
15. Mercadante S. Oral mor phine con sump tion in It aly and Sic -
cacy and qual ity of life. J Pain Symp tom Man age 1997;
ily. J Pain Symp tom Man age 1998;14:227-30.
13:254-61.
16. Schug SA, Zech D, Grond S, Jung H, Meuser T, Stobbe B. A
21. Haazen L, Noorduin H, Megens A, Meert T. The con sti pa -
long term sur vey of mor phine in can cer pain pa tients. J Pain
tion-inducing po ten tial of mor phine and trans derm al
Symp tom Man age 1992;7:259-66.
fentanyl. Eur J Pain 1999;3:9-15.
