1520
O’Ryan et al.
JID 2000;182 (November)
Table 1. Human calicivirus (HuCV) positivity by year and by source of stool samples.
No. of HuCV-positive samples/
no. of samples tested (%), by year
Age in months,
Source
mean (range)
1997
1998
1999
Total
Diarrhea patients
Emergency department
Outpatient clinic 1
Outpatient clinic 2
Hospital 1
13.7 (0.5–45)
17 (1–58)
20.5 (0.5–60)
7.6 (0.5–30)
9.5 (1–48)
8/39 (21)a
3/25 (12)
Not done
1/12 (8)
18/120 (15)
0/42
11/98 (11)
5/30 (17)
2/61 (3)
0/89
0/13
4/96 (4)
0/18
1/41 (2)
5/257 (2)
26/248 (10)
3/80 (4)
15/194 (8)
6/60 (10)
3/102 (3)
53/684 (8)
Hospital 2
Total
Not done
15.2 (0.5–60)
12/76 (16)a
36/351 (10)
Control patients
Emergency department
Hospital 1
Hospital 2
Total
11.3 (0.5–40)
6.7 (1–12)
8.9 (0.7–52)
9 (0.5–52)
Not done
Not done
Not done
Not done
0/43
0/11
0/11
0/65
1/37 (3)
0/7
0/21
1/80 (1)
0/18
0/32
1/65 (2)
1/130 (!1)
a
P ! .001 by x2 test for trend of decrease in detection rates observed, 1997–1999.
health care facilities, including 2 outpatient clinics, a large ED, and
2 hospital wards of the Chilean National Health Care System,
enrolled patients. Combined, these sites serve a population of ∼2
million persons, the vast majority of whom are of low or mid-low
socioeconomic status. At each site, all children р60 months of age
with acute, nonbloody diarrhea of !3 days’ duration were eligible
for stool sample collection. Three sites participated from 1997
through 1999 and 2 from 1998 through 1999. Both participating
outpatient clinics and 1 hospital ward (Roberto del R´ıo Hospital;
hospital 2) belong to the north health care area of Santiago. During
1998–1999, they had means of 1200 pediatric outpatient visits and
190 pediatric hospitalizations per month. The participating ED and
the other hospital ward are part of the So´tero del R´ıo Hospital
(hospital 1) in southeast Santiago. These had means of 9200 ED
visits and 311 hospitalizations for children per month during a
similar period. Designated personnel obtained samples in the out-
patient clinics and ED during weekday regular work hours. A stool
sample was obtained by hospital staff from all children hospitalized
with acute nonbloody diarrhea, as part of the routine workup for
severe acute diarrhea.
Sample processing. All stool samples were fresh (!30 min old)
and were stored at Ϫ70ЊC until testing. Selected samples were pro-
cessed for HuCVs at the University of Chile microbiology labo-
ratory. Samples were processed for RT-PCR by the TRIzol ex-
traction method (Gibco BRL, Gaithersburg, MD). A 500-mL
suspension of stool sample (1:5 in PBS and vortexed with 150 mL
of Genetron; Sigma, St. Louis) was centrifuged for 5 min at 18,000
g. A mix that included 200 mL of the supernatant, 600 mL of
TRIzol-R, and 160 mL of chloroform was incubated for 3 min at
room temperature and centrifuged at 13,000 rpm for 15 min at
4ЊC. Aqueous-phase RNA was extracted with isopropanol over 12
h at Ϫ20ЊC and then precipitated with ethanol. The precipitant
was resuspended in 20 mL of RNAse- and DNAse-free distilled
water [10].
free distilled water). The reaction mixture was incubated for 1 h
at 42ЊC. After this process, the tubes were spun, and 50 mL of PCR
mix containing 5 mL of 10ϫ buffer, 2 mg of primer 290, 5 U of
Taq polymerase, and 42 mL of RNAse- and DNAse-free distilled
water was added. Samples were denatured for 3 min at 96ЊC and
were subjected to 40 cycles of amplification: 30 s at 96ЊC, 1 min
at 50ЊC, and 2 min at 72ЊC. The RT-PCR product was separated
on a 1% agarose gel, stained with ethidium bromide, and visualized
by UV light.
Results
Population enrolled and selection of samples for HuCV test-
ing. In total, 1787 diarrhea samples were obtained from ED
patients during the 31-month study period. For HuCV testing,
we selected 10 samples per month or all available ED samples
if 10 were not available, for a total of 248 samples (table 1).
In addition, we tested 80 samples from age-matched control
children without diarrhea who presented to the ED from April
1998 through August 1999. Of the diarrhea samples from hos-
pitalized children, 280 were from hospital 1 and 300 from hos-
pital 2 during the 25- and 17-month enrollment periods, re-
spectively. For HuCV testing, we randomly selected 3 and 6
samples per month, respectively (or all available): 60 samples
from hospital 1 and 102 from hospital 2. In addition, we tested
65 samples obtained from time- and age-matched children who
were hospitalized during 1998 and 1999 for reasons other than
diarrhea. We tested all 274 diarrhea samples from the 2 par-
ticipating outpatient clinics for HuCVs. Control samples were
not available from the outpatient clinics. As expected, children
hospitalized for management of acute diarrhea were, in general,
younger than children managed in the ED; the latter were
younger than children managed in outpatient clinics (table 1).
Characterization of HuCV positivity. In total, 53 (8%) of
684 diarrhea stool samples were positive for HuCV (table 1).
During 1998–1999, 41 (6.7%) of 608 diarrhea samples and 1
(0.8%) of 130 control samples were positive for HuCV (P p
.01, x2 test with Yates’s correction). HuCV-positive samples
were obtained from all sites throughout the study period. There
Primers for RT-PCR 289/290 produce 319 and 331 nt products
for NV-like and Sapporo-like HuCVs, respectively, in the 3D ge-
nomic region of open-reading frame 1, a relatively conserved region
of the genome that encodes for the polymerase. For RT-PCR, 3
mL of extracted RNA was mixed with 48 mL of reaction mixture
(8 U of avian myeloblastosis virus RT, 20 mM of each dNTP, 0.2
mg of primer 289, 10 U of RNAsin, 5 mL of 1% bovine serum
albumin, 5 mL of 10ϫ buffer, and 31 mL of RNAse- and DNAse-