| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Medical Imaging, University Health Network and Mount Sinai
Hospital, Toronto, ON, Canada.
2 Princess Margaret Hospital, 610 University Ave., Office 3-956, Toronto, ON M5G
2M9, Canada.
3 Department of Medicine, University Health Network, Toronto, ON, Canada.
4 Department of Medicine, Mount Sinai Hospital, Toronto, ON, Canada.
5 Department of Medicine, University of Toronto, Toronto, ON, Canada.
6 Department of Medical Imaging, The Scarborough Grace Hospital, Scarborough,
ON, Canada.
7 Department of Pulmonary and Critical Care Medicine, York Central Hospital
Richmond Hill, ON, Canada.
Received June 9, 2003;
accepted after revision August 7, 2003.
Address correspondence to N. S. Paul.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. A retrospective review of 439 radiographs was performed for 51 patients with a final diagnosis of probable SARS. Forty-nine patients were followed up for a mean interval of 23 days (range, 2–63 days).
RESULTS. Abnormal findings on a chest radiograph were noted at presentation in 80.4% (41/51) of patients. Four radiographic patterns were seen: normal (group 1) in 19.6% (10/51), focal opacity (group 2) in 39.2% (20/51), multifocal opacities (group 3) in 27.5% (14/51), and diffuse air-space opacification (group 4) in 13.7% (7/51). Radiographic progression of disease occurred in 38.8% (19/49) of the patients in groups 1–4. There were no deaths in groups 1 and 2. In group 3, one (7.7%) of the 13 patients died. Five (71.4%) of the seven patients in group 4 died. Overall, 12.2% (6/49) of the patients died, all of whom had diffuse air-space opacification on the last chest radiograph. In these patients, medical comorbidity was present in 66.7% (4/6), and the exposure history was known in 83.3% (5/6). Death occurred at a mean interval of 18.2 days (range, 9–36 days) from the initial exposure.
CONCLUSION. Patients presenting with normal findings or focal air-space opacity on chest radiographs had a good clinical outcome. Patients with multifocal opacities that progressed to diffuse air-space opacification and patients presenting with diffuse air-space opacification had a high fatality rate, but patients in this group were also older and more likely to have comorbid conditions. Patients with SARS present with recognizable patterns of disease that have prognostic significance.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although highly contagious, the virus exhibits variable pathogenicity; in some patients even casual exposure is sufficient to cause symptoms, whereas others appear unaffected even by extensive exposure [5]. Several radiographic patterns have been noted on initial chest radiography or on thoracic CT [2, 5–9]. The prognostic significance of these patterns has not been elucidated, and there is little information about follow-up of these patients. A recent retrospective study [10] addressed the pattern of radiographic disease progression in patients with SARS but not the significance of the radiographic appearance at presentation. The purpose of this study was to assess the prognostic significance of the initial radiographic pattern of disease by relating this pattern to clinical outcome.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients
In this retrospective study, all the patients with a diagnosis of probable
SARS admitted to two major University of Toronto teaching hospitals (sites 1
and 2) and a local community hospital (site 3) from March 7 to April 12, 2003,
were recruited. The patients were identified from a SARS registry compiled by
the Greater Toronto Area SARS Study Group
[6] and from participating
physicians at the admitting hospitals involved in the study. The SARS registry
consisted of patients admitted from March 7 to April 10, 2003. Epidemiologic
data documenting the onset of symptoms suggestive of SARS were available from
the Greater Toronto Area SARS Study Group for 66.7% (34/51) of the patients in
this study. The patients were classified according to the SARS case
definitions established by the World Health Organization
[11].
A suspect case of SARS is a person presenting after November 1, 2002, with a history of high fever (> 38°C) and cough or breathing difficulty and one or more of the following exposures during the 10 days before the onset of symptoms: close contact with a person who has a suspect or probable case of SARS, history of travel to an area with recent local transmission of SARS, or residence in an area with recent local transmission of SARS. A suspect case of SARS is a person with an unexplained acute respiratory illness resulting in death after November 1, 2002, but on whom no autopsy was performed and one or more of the following exposures during the 10 days before the onset of symptoms: close contact with a person with a suspect or probable case of SARS, history of travel to an area with recent local transmission of SARS, or residence in an area with recent local transmission of SARS.
A probable case is a suspect case of SARS with radiographic evidence of infiltrates consistent with pneumonia or respiratory distress syndrome on chest radiography, a suspect case that is positive for SARS coronavirus by one or more assays, or a suspect case with autopsy findings consistent with the pathology of respiratory distress syndrome without an identifiable cause.
In our series, the diagnoses for all patients were probable SARS rather than suspect SARS based on radiographic evidence of pneumonia or respiratory distress syndrome.
Data
All the examinations at sites 1 and 2 were soft-copy chest radiographs
interpreted from a workstation (eFilm 1.8, eFilm/Merge Technology, Toronto,
ON, Canada), and the radiographs obtained at site 3 were conventional
radiographs. Forty-three (84.3%) of the 51 initial chest radiographs were
obtained as portable examinations; all the in-patient studies were single-view
portable frontal radiographs. All examinations were interpreted by consensus
of three experienced chest radiologists who knew that the patients were
suspect SARS cases but were unaware of other clinical details.
The final diagnosis was probable SARS in 51 patients: 16 men and 35 women with a mean age of 54.2 years (range, 21–88 years). Reviewers interpreted 439 radiographs, with a mean of 8.6 radiographs per patient (range, 1–24). The mean followup was 20.4 days (range, 1–55 days).
The initial chest radiograph was graded as normal or abnormal. Patients in whom the initial chest radiographs were normal but subsequently developed air-space opacities on chest radiography or CT were included as probable cases of SARS (group 1). Abnormal radiographic patterns were described as focal (group 2), multifocal (group 3), or diffuse (group 4) opacities. Focal opacities were noted as isolated areas of consolidation. Multifocal opacities were defined as several discrete but separate areas of consolidation; these findings were further classified as unilateral or bilateral. Diffuse opacities were defined as widespread air-space consolidation without discrete lesions. Each opacity was further described by anatomic location as predominantly upper zone, middle zone, or lower zone and as peripheral or central. The upper zone was demarcated by the area from the lung apex to the anterior aspect of the second rib; the middle zone, by the anterior aspects of the second to fourth ribs; and the lower zone, by the remaining lung. A central location was ascribed if the opacity was predominantly in the medial half of the zone and a peripheral location if predominantly in the lateral half.
Additional features such as lymphadenopathy, pleural effusions, and pneumothorax were noted. The extent of radiographic disease was further assessed on the series of follow-up chest radiographs with particular attention to deterioration in the radiographic appearance (progression from normal to focal, focal to multifocal, or multifocal to diffuse opacities), timing of the peak radiographic abnormality, and pattern of disease on the last available radiograph. The peak radiographic abnormality was assessed qualitatively by reviewing the radiographic series for each patient and identifying the radiograph that showed either the largest area of lung involved (for focal or diffuse opacities) or the greatest number of zones affected (for multifocal opacities).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Patients who presented with a normal radiograph had a mean interval of 6.5 days (2–12 days) between exposure to SARS and hospital admission compared with an interval of 10.5 days (4–29 days) for the other patients (Table 2). Sixty-nine percent (9/13) of patients with multifocal disease had bilateral disease.
|
The pattern of zonal involvement was as follows: right upper zone, 19.6% (10/51); right middle zone, 52.9% (27/51); right lower zone, 43.1% (22/51); left upper zone, 19.6% (10/51); left middle zone, 39.2% (20/51); and left lower zone, 43% (22/51). The right lung was involved in 78.4% (40/51) and the left lung in 58.8% (30/51) of patients.
Of the 51 initial radiographic examinations, 10 (19.6%) showed normal findings, and seven (13.7%) revealed diffuse air-space opacification. On the examinations for the remaining 34 patients, a predominantly peripheral pattern of disease was noted in 82.4% (28/34) and a predominantly central pattern in 17.6% (6/34). Lymphadenopathy and pleural disease (pleural effusions or pneumothorax) were not seen.
Progression
Table 3 shows the data
regarding the radiologic progression of disease.
|
Normal radiographic findings.—Of the 10 patients who presented with normal radiographic findings, one patient had persistently negative radiographic findings despite a series of 11 examinations over an interval of 19 days. In this patient, CT of the thorax showed subtle lung opacities. Because the radiographic series remained normal throughout, this patient was excluded from further analysis and is the subject of a case report [12]. The remaining nine patients developed lung opacities at a mean interval of 3.4 days (range, 1–6 days), with two patients having a persistently normal radiograph for 6 days from initial presentation. The peak radiographic abnormality occurred at a mean interval of 6.9 days (range, 1–22 days) from initial presentation.
Focal opacity.—This pattern of disease was the initial finding in 39.2% (20/51) of patients (Fig. 1). The peak abnormality occurred at a mean interval of 5.9 days (range, 1–25 days) from initial presentation. In 35% (7/20) of the patients, the chest radiograph at presentation showed the most severe radiologic change and subsequent radiographs showed progressive improvement. One radiographic pattern noted in this group of patients was an initial growth in volume of the focal opacity (Fig. 2A, 2B), which was followed by a subsequent decrease in volume and an increase in radiographic density until a radiograph showed either normal findings or a residual linear opacity (Fig. 3A, 3B).
|
|
|
|
|
Multifocal opacities.—This pattern of radiographic disease was seen in 27.5% (14/51) of patients. One patient who presented with bilateral air-space opacities in the middle zone was subsequently transferred to another hospital. Because follow-up radiographs were not available for this patient, she was excluded from further analysis. In two (15.4%) of 13 patients, disease progressed from unilateral to bilateral (Fig. 4A, 4B); in another two patients (15.4%), disease progressed from multifocal to diffuse. The peak radiographic abnormality occurred at a mean interval of 5.3 days (range, 1–18 days) from presentation.
|
|
Multifocal bilateral air-space opacities were noted to appear and disappear in different lung zones over the course of several days in a "wax-and-wane" or sinusoidal pattern in 23.1% (3/13) of the patients in this group.
Diffuse air-space opacification.—This pattern of radiographic disease was seen in 13.7% (7/51) of patients. Widespread air-space opacification was the feature in this group of patients. (Fig. 5). Four (57%) of seven patients presented with peak radiographic disease, whereas three (42.9%) of seven patients showed peak disease as extensive air-space opacification at a mean of 2.3 days (range, 1–9 days) after presentation.
|
Outcomes
The outcomes for the different groups are shown in
Table 4. Overall, 53% (26/49)
of patients had normal findings on chest radiography at a mean interval
follow-up of 14.6 days (range, 1–38 days). Residual radiographic
abnormalities ranged from subtle linear opacities to multifocal bilateral
air-space disease. There were no deaths in groups 1 and 2. The death in group
3 was of a previously healthy 73-year-old woman who presented with unilateral
multifocal opacities that progressed to diffuse air-space opacification within
18 days. Among group 4 patients, 80% (4/5) of the deaths were associated with
comorbid medical conditions (diabetes mellitus, carcinoma, or cardiac
disease). One of the fatalities was a liver transplant recipient.
|
Death occurred in 12.2% (6/49) of the patients overall. The patients who died were predominantly men (66.7%; men–women ratio, 4:2) with a mean age of 68 years (range, 44–80 years).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, patients with a final clinical diagnosis of probable SARS presented with one of four radiographic patterns. Patients with normal findings or with a focal opacity on chest radiography at initial presentation were predominantly young women who became symptomatic at a mean interval of 4.8 days (range, 2–13 days) after exposure. Although the radiographic pattern of multifocal opacities progressed in 31% (9/29) of these patients, they had a good clinical outcome.
Patients with multifocal opacities were older than those in other groups and were predominantly women. The time from first exposure to the onset of symptoms was longer at a mean interval of 7.8 days (range, 3–10 days), and radiographic disease was progressive in 31% (4/13) of these patients. In this group, a sinusoidal pattern of radiographic disease was observed in 23.1% (3/13). The fatality in this group was a previously well 73-year-old woman who presented 17 days after exposure and developed diffuse air-space opacification after an initial radiograph showed unilateral multifocal opacities. She died 36 days after SARS exposure.
The other radiographic pattern observed was diffuse air-space opacification. Compared with patients in other groups, the patients in this group were predominantly older men with a shorter time from first exposure to the onset of symptoms, a mean interval of 3.3 days (range, 2–5 days). There was a high incidence of comorbidity in this group, and the fatality rate of 71.4% (5/7) was highest in these patients. Death occurred after a mean interval of 13.8 days (range, 9–17 days) after SARS exposure. Overall, six (12.2%) of the 49 patients in this cohort died.
The main limitation of this study is the reliance on portable chest radiographs. The diagnostic quality of a portable chest radiograph is greatly dependent on the patient body habitus and positioning at the time of exposure. Even a small degree of patient rotation, particularly in female patients with substantial breast tissue, can cause considerable difficulty in excluding peripheral lung infiltrates. The pattern in probable SARS patients was seen principally in the middle and lower zones and was peripheral in location. CT is a more sensitive technique for depicting lung consolidation as was shown in one of our patients with positive findings on CT and a series of normal chest radiographs. Despite the limitations of portable chest radiography, portable examinations were performed to limit the transfer of patients with SARS in the hospital.
All patients with suspect or probable SARS were placed in negative-pressure isolation rooms or single-patient rooms to contain the spread of infection in the health care setting. The spread of SARS is believed to be through large respiratory droplets and direct or indirect contact with respiratory secretions. Respiratory precautions that include the use of masks, goggles, and face shields and contact precautions that include the use of gloves and gowns are implemented as standard procedures for infection control in the management of these patients.
There is no treatment that is of proven benefit for the management of patients with SARS. Most patients with SARS receive empiric antibiotic therapy consistent with treatment guidelines for the management of community-acquired pneumonia [13]. During this outbreak of SARS in the Greater Toronto Area, the majority of the patients (88%) received treatment with ribavirin, an antiviral agent with a broad spectrum of activity against RNA viruses including coronaviruses. This therapy was associated with a high incidence of adverse effects without proven benefit [6]. During the second outbreak of SARS in Toronto, ribavirin was not used for treatment (Gold W, personal communication). Corticosteroid therapy according to various dosing regimens was also prescribed in a significant proportion of patients [6]. The theoretic benefit of steroids is a reduction in lung inflammation, a treatment strategy that has been used in patients with AIDS who have severe Pneumocystis carinii pneumonia [14].
In this study, the patients who presented with the most extensive radiographic changes (groups 3 and 4) had the longest delay between symptom development and hospital admission. Patients with normal findings or with a focal opacity on radiography at presentation were admitted at a mean of 3.6 days after developing symptoms, whereas patients with multifocal and diffuse disease presented at a mean of 5.6 days. Allowing for the higher incidence of comorbidity in groups 3 and 4, it is possible that the earlier hospital admission and subsequent initiation of treatment in groups 1 and 2 contributed to the more benign clinical course in these patients.
In conclusion, SARS is a highly contagious viral pneumonia that has a high fatality rate in older patients with medical comorbidity. In this study, we observed that the initial pattern of radiographic presentation in patients with probable SARS might be helpful in predicting disease progression and outcome. Patients with normal findings or a focal opacity on the initial radiograph had a good clinical outcome. Patients with multifocal opacities or with diffuse opacities had a worse clinical outcome; all the patients who died were from these two groups. However, the patients with multifocal and diffuse opacities were also older and more likely to have comorbid conditions. We also postulate that earlier hospital admission and initiation of corticosteroid therapy may prevent the progression to a more extensive radiographic pattern of disease.
Acknowledgments
We thank Alan Detsky and the Greater Toronto Area SARS Study Group for use
of their epidemiologic data.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
