ID Journal Club

1
ID Journal Club

• Hrach Ike Kasaryan

2
Article

• Do Follow-Up Imaging Examinations Provide Useful
  Prognostic Information in Patients with Spine
  Infection?
• Clinical Infectious Diseases    200643172-179

3
Introduction

• There has been a shift from primarily surgical
  therapy toward primarily medical therapyin
  certain spinal infections
• Imaging studies such as CT/MRI are used to follow
  these patients with no documentation if they
  actually help or not
• AND COST A LOT OF MONEY!

4
Purpose

• to determine whether applying a simple grading
  system focused on soft-tissue findings for
  follow-up imaging results 48 weeks after
  initiation of therapy in patients with spine
  infection provides clinically useful prognostic
  information

5
Methods

• Single Center, RETROSPECTIVE
• Found patients through their electronic data base
• Patients from 1994-2002
• Patients who met criteria, and had available
  follow up MRI/CT at 4-8 weeks
• 79 patients found to fit criteria for study

6
(No Transcript)
7
(No Transcript)
8
EXCLUSIONS

• tuberculous or brucellosis was excluded
• patients who had spinal instrumentation present
  at the time of diagnosis

9
Methods.

• An ID physician than reviewed the MRI/CT at 4-8
  weeks on these cases that fit (79) and assinged a
  simple grading system (improved, equivical, or
  worst) see table 1
• They than compared that to the clinical assesmant
  based on a three category grade system based on
  pain medication taken

10
Results
11
Results

• None of the 27 patients who demonstrated
  improvement on follow-up images developed
  microbiologic failure, compared with 5 (13) of
  38 patients with equivocal follow-up imaging
  findings and 4 (29) of 14 patients whose
  follow-up imaging findings worsened

12
Results

• Twenty-six (96) of 27 patients with improvement
  in the imaging findings also experienced clinical
  improvement at follow-up
• the need for unexpected surgery was more commonly
  seen among patients with imaging findings that
  were equivocal or worst (table 5)
• The estimated 1-year rates of survival free of
  clinical failure were 100, 78 (95 CI,
  6289), and 29 (95 CI, 1156) for improved,
  equivocal, and worse
• The estimated 1-year rates of survival free of
  microbiologically confirmed failure were 100,
  89 (95 CI, 7496), and 56 (95 CI, 2483)
  for improved, equivocal, and worse

13
REsults

• Three (6) of 52 patients who had clinical
  improvement at follow-up developed microbiologic
  treatment failure, compared with 4 (21) of 19
  patients who had equivocal clinical status and 2
  (25) of 8 patients who had worse clinical
  status.
• Of the 3 patients who experienced clinical
  improvement and who developed microbiologic
  treatment failure, 2 had markedly elevated levels
  of inflammatory biomarkers during the 48-week
  follow-up period, identifying them as high risk
  for failure

14
Study showed imaging does make a difference

• BUT WAIT!!

15
discussion

• So they looked at biomarkers of inflammation and
  realized it could be a way to screen patients if
  they needed to be imaged.
• They found that people who had elevated
  biomarkers at 4-8 weeks were more likely to have
  clinical/microbiologic/radiologic failure

16
discussion

• They suggested to get a CRP/ESR at baseline
• Follow up in 4-8 weeks and If the ESR/CRP is
  higher than move on to the imaging study, if it
  is the same or better, there is less of a chance
  of finding something
• But there is the few who would have no change in
  ESR/CRP who would need intervention
• But with regards to cost, this would be the most
  cost effective way
• THIS IS ONLY THEIR SUGGESTION!!

17
PROs

• Largest study of its type
• Good amount of data available in all cases
• SIMPLE GRADING SYSTEM
• Good follow up

18
Cons

• RETROSPECTIVE
• Selection bias -gt patients who need imaging may
  have been the sicker patients than those who
  didnt
• CLINICAL IMPROVEMENT definitions based soley on
  pain medications which may also be misleading

19
Summary

• Imaging the spine does help theraputic decisions
  and patient outcome
• A possible cost effective algorythm may be used
• A randomized trial is needed to confirm
• WOULD YOU IMAGE EVERYONE???