This document summarizes the STICH II trial which compared early surgery versus initial conservative treatment for spontaneous supratentorial lobar intracerebral haemorrhages between 10-100mL. The trial involved 601 patients randomized to either early surgery within 12 hours or initial conservative treatment with delayed surgery if needed. The primary outcome of death or severe disability at 6 months showed no significant difference between the groups. Secondary outcomes of mortality, functional scales also showed no significant differences, indicating that early surgery did not improve outcomes over initial conservative treatment for these types of hemorrhages.

1. Dr. Gaurav Kumar
Emergency Medicine Resident
Aiims
9/27/2014 1

2. Background
STICH –pontaneous supratentorial lobar intracerebral
haematomas.
9/27/2014 2

3. Background
Spontaneous intracerebral haemorrhage (ICH) accounts for 10 to
40% of all cases of stroke
The morbidity and mortality exceed 60%.
Young disabled survivors are a significant burden to both Health
Services and family.
9/27/2014 3

4. Background
Within the spectrum of spontaneous intracerebral haemorrhage there
are some patients with large or space occupying haemorrhage who
require surgery for neurological deterioration .
 And others with small haematomas who should be managed
conservatively.
There is equipoise about the management of patients between these
two extremes.
9/27/2014 4

5. Findings of previous trials
The first randomised trial of Surgical Treatment of ICH, published in
1961.
But it did not show a significant advantage for either surgical or
conservative treatment.
Many other studies were done after this but none showed benefit of
one over other method of Rx.
9/27/2014 5

6. STICH I
The need to gain evidence to support clinical decision making led to
the initiation of the Surgical Trial in Intracerebral Haemorrhage
(STICH).
Started in 1998, this trial is the largest to date and successfully
recruited 1033 patients from 87 centres around the world.
9/27/2014 6

7. RESULT OF STICH I
 Showed a small non-significant advantage for surgery (Mendelow et
al. 2005)
9/27/2014 7

8. UNFORTUNATE OUTCOME OF STICH I
Had been that many people misinterpreted the
results to argue that there was no need to operate on
patients with ICH at all !
9/27/2014 8

9. Neurosurgeons know that early removal of an intracranial
haemorrhage is highly effective in the context of trauma.
 It seems unlikely that surgery would be of benefit in one scenario and
not in the other.
Hence the STICH II Trial.
9/27/2014 9

10. TRIAL DESIGN
Prospective, randomized, parallel group trial.
In all, 129 NSx units in 39 countries.
 Only patients for whom the treating neurosurgeon is in equipoise
about the benefits of early craniotomy compared to initial conservative
treatment are eligible for the trial.
 Outcome is measured at six months via a postal questionnaire
9/27/2014 10

11. INCLUSION CRITERIA
All of the following :
 Evidence of a spontaneous lobar ICH on CT scan (1 cm or less from
the cortex surface of the brain)
Patient within 48 hours of ictus.
 Best MOTOR score on the Glasgow Coma Scale (GCS) of 5 or 6 and
best EYE score on the GCS of 2 or more
Volume of haematoma between 10 and 100 ml [Calculated using
(a × b × c)/2 method].
9/27/2014 11

12. Eligible CT
9/27/2014 12

13. EXCLUSION CRITERIA
Any of them :
Clear evidence that the haemorrhage is due to an aneurysm or
angiographically proven arteriovenous malformation.
Intraventricular haemorrhage of any sort.
ICH secondary to tumour or trauma.
9/27/2014 13

14. Not CT
9/27/2014 14

15. EXCLUSION CRITERIA
Basal ganglia, thalamic, cerebellar or brainstem haemorrhage or
extension of a lobar haemorrhage into any of these regions
If surgery cannot be performed within 12 hours.
 Patients who did not consent to participate
9/27/2014 15

16. INTERVENTIONS
 Patients were randomly allocated to either early surgery or initial
conservative treatment.
All patient must have an baseline CT head.
In the early surgery group, surgeons were expected to undertake
evacuation of the haematomas within 12 h.
 In the initial conservative treatment group, delayed evacuation was
permitted if judged clinically appropriate.
9/27/2014 16

17. RANDOMIZATION
It is not possible to blind either patients or treating surgeons to when
the patient has had surgery or whether they have had surgery.
To minimise possible sources of bias randomisation is undertaken
centrally by an independent organization.
Centre for Healthcare Randomised Trials, Aberdeen.
9/27/2014 17

18. RANDOMIZATION
Randomisation must take place within 48 hours of ictus
Eligible patients randomly assigned in a 1:1 ratio to early surgery or
early medical treatment group.
Randomization done with the use of telephone and internet based
system with stratification according to country and planned operation.
9/27/2014 18

19. STUDY END POINTS
Primary end point :
 DEATH.
Severe disability.
9/27/2014 19

20. STUDY END POINTS
Secondary end point :
mortality,
 time to death,
prognosis-based dichotomised Rankin and GOSE, and Rankin
and EuroQol; all measured at 6 months.
9/27/2014 20

21. Statistical analysis
 Analysis will be on an “intention to treat” basis
 a sample size of 566 (283 in each group) was needed to show a 12%
benefit from surgery (two-sided significance level of 0·05) with 80%
power.
Hence a sample size of 600 patients to allow for withdrawals and
crossovers.
9/27/2014 21

22. Statistical analysis
Primary outcome analysis was a simple categorical frequency
comparison by use of the χ2 test for prognosis-based favourable and
unfavourable outcome on GOSE(GLASGOW OUTCOME SCALE-EXTENDED)
Secondary analysis consisted of a Kaplan-Meier survival curve with
log-rank test, χ2 test for mortality at 6 months, and 6 month prognosis-based
Rankin
9/27/2014 22

23. RESULTS
9/27/2014 23

24. No. of patients in study=601
294
307
no. of patients
conervaticve gp. surgery gp.
9/27/2014 24

25. STUDY OUTLINE

26. The two groups were well matched at baseline,
57% of total were male and median age was 65.
50% of patient in early surgery group and 49% in early conservative
group had a GCS of 14 or 15 at the time of randomization.
9/27/2014 26

27. Characteristics of haematomas
EARLY SURGERY GROUP CONSERVATIVE GROUP
VOLUME(ML)
MEDIAN 38 36
MEAN 41.4 41.0
DEPTH(MM)
MEDIAN 1 1
MEAN 1.6 1.6
SIDE OF H’AGE
LEFT 158 (52%) 149 (51%)
RIGHT 147 (48%) 142 (49%)
9/27/2014 27

28. outcome
Early surgery group Early conservative group
9/27/2014 28
Primary outcome
Prognosis based
Unfavourable 174(59%) 178(62%)
Favourable 123(41%) 108(38%)
Secondary outcome
Mortality at 6 months
Dead 54(18%) 69(24%)
Alive 244(82%) 222(76%)
Prognosis based modified Rankin
Unfavourable 155 (53%) 158 (56%)
Favourable 140 (47%) 126 (43%)

29. SUBGROUP ANALYSIS

30. KAPLAN-MEIER SURVIVAL CURVE
Source: The Lancet 2013; 382:397-408 (DOI:10.1016/S0140-6736(13)60986-1)

31. Discussion
 DID not find significant evidence to support that early surgery
compared with initial conservative treatment (with delayed surgery if
the patient deteriorates) improves outcome in conscious patients in
whom there is a superficial intracerebral haemorrhage of 10—100 mL
and no evidence of intraventricular haemorrhage.
9/27/2014 31

32. Strengths of the study
Prospective randomized trial.
Baseline characteristics of patients similar.
Early randomization and early surgery.
9/27/2014 32

33. Limitations of the study
Cross over : 62 (21%) of 291 patients assigned to initial conservative
treatment went on to have delayed surgery.But because of the
intention-to-treat analysis they remained in the initial conservative
treatment group.
9/27/2014 33

34. Limitations of the study
More than half of the patients had good prognosis(i.e. 66% pt inearly
surgery group and 64% pt inconsevative group)
But the patients who benefited most from surgery were poor
prognosis group.
9/27/2014 34

35. Critical Appraisal
Was the randomization list concealed ? yes
Was the follow up of the patients sufficiently long and complete ? yes
Were the patients, clinicians and the study personnel kept “blind” to treatment ? No
Were the groups treated equally ? yes
Were the groups similar at the start of the trial ? yes
Does these results apply to our patient ? ?
9/27/2014 35

36. SUBGROUP ANALYSIS