GFR talk

Document Sample
GFR talk Powered By Docstoc
					                            East Kent Hospitals
                                        NHS Trust




  eGFR – Laboratory Issues

               Renal Association –
               Clinical Services Meeting

               Dr Edmund Lamb
Ucr x V        4th October 2005
Pcr x T
                                    East Kent Hospitals
                                                NHS Trust




                  Overview
•   eGFR using formulae
•   Creatinine measurement
•   Creatinine standardisation – effect on eGFR
•   Solutions and further work?
                                       East Kent Hospitals
                                                   NHS Trust




                    Givens
• GFR is the best overall index of kidney function
• Gold-standard GFR techniques are not practical
  for the entire CKD population
• Estimates of GFR are better/more practical than
  creatinine clearance
• Estimates of GFR are more sensitive for CKD
  than creatinine alone
                                             East Kent Hospitals
                                                         NHS Trust




                Estimated GFR
Equations based on serum creatinine but taking into account
  non-renal influences improve its relationship with GFR

“GFR should be estimated from prediction equations that
  take into account serum creatinine and some or all of age,
  gender, race and body size…….In adults, either
  Cockcroft & Gault or MDRD equations are useful”

NKF-KDOQI 2002
                                            East Kent Hospitals
                                                        NHS Trust




   Renal NSF Part Two: Chronic Kidney
  Disease, Acute Renal Failure and End of
         Life Care (February 2005)
• Quality Requirement One: Prevention and Early
  Detection of CKD

• “Local health organisations can work with pathology
  services and networks to develop protocols for measuring
  kidney function by serum creatinine concentration
  together with a formula-based estimation of GFR,
  calculated and reported automatically by all clinical
  biochemistry laboratories”.
             East Kent Hospitals
                         NHS Trust




Which Formula and
      When?
                                                                     East Kent Hospitals

             GFR prediction equations                                            NHS Trust




Cockroft-       (140 - age) x weight/72 x Scr (x 0.85 if female)
Gault
MDRD 1          170 x Scr-0.999 x age-0.176 x (0.762 if female) x (1.180 if black) xSu-0.170
                x Alb+0.318
MDRD 2          186 x Scr-1.154 x age-0.203 x (1.212 if black) x (0.742 if female)

Jelliffe 1      98 - 0.8 x (age - 20)/Scr (x 0.90 if female)
Jelliffe 2      Male: 100/Scr - 12
                Female: 80/Scr - 7
Mawer           Male: weight x [29.3 - (0.203 x age)] x [1 - (0.03 x Scr)]/(14.4 x Scr) x
                (70/weight)
                Female: weight x [25.3 - (0.175 x age)] x [1 - (0.03 x Scr)]/(14.4 x Scr)
                x (70/weight)
Bjornsson       Male: [27 - (0.173 x age)] x weight x 0.07/Scr
                Female: [25 - (0.175 x age)] x weight x 0.07/Scr
Gates           Male: (89.4 xScr-1.2) + (55 - age) x (0.447 xScr-1.1)
                Female: (60 xScr-1.1) + (56 - age) x (0.3 xScr-1.1)
                                             East Kent Hospitals
                                                         NHS Trust




          MDRD formula (2000)
Abbreviated MDRD (4-v)
186 x [serum creatinine (umol/L) x 0.011312]-1.154 x age-
  0.203 x (1.212 if black) x (0.742 if F)



Published in abstract form only
No requirement for urea and albumin (or weight)
Little loss of accuracy R2 = 0.89, bias -12.1%
90% of subjects within 30% of true GFR
                                   East Kent Hospitals
                                               NHS Trust


         National/International
          Recommendations
• NKF – now favour 4v MDRD
• NKDEP – 4v MDRD
• ERA EBPG – any method validated against BSA-
  corrected GFR
• KDIGO - method validated against a gold-
  standard measure of GFR
• Kidney Health Australia – 4v MDRD
• UK (RCPhys/RA/RCGP/ACB/DoH) – 4v MDRD
                                                             East Kent Hospitals
                                                                         NHS Trust




                       C&G vs MDRD
Lewis et al 2001               African-Americans
Vervoort et al 2002            Type 1 diabetes
Bostom et al 2002              CKD patients
Lin et al 2003                 Healthy donors
Rodrigo et al 2003             Transplant recipients
Pierrat et al 2003             Children >12 y
Van den Noortgate et al 2003   „Old old‟
Lamb et al 2003                Older people
Rule et al 2004                Healthy donors
Poggio et al 2005              CKD patients
Poggio et al 2005              Kidney donors
Froissart et al 2005           European CKD

•   Reviewed Lamb, Tomson, Roderick. Ann Clin Biochem 2005 (elamb@nhs.net)
•   Cochrane review in progress
•   4v-MDRD offers practical advantages and is more accurate and precise for stage 3-5
    CKD
                                                   East Kent Hospitals
                                                               NHS Trust




         MDRD and level of GFR
                       CKD                  ‘healthy’
Rule et al 2004        -6.2%                -29%
Poggio et al 2005      -0.5 mL/min/1.73 m2 - 9.0 mL/min/1.73 m2

Froissart et al 2005   1.3 mL/min/1.73 m2   -3.3 mL/min/1.73 m2


   •NKDEP/KHA – don’t report if >60 mL/min/1.73 m2
   •UK CKD – don’t report if >90 mL/min/1.73 m2
                                          East Kent Hospitals
                                                      NHS Trust




Froissart et al 2005: Scatter increases as GFR
approaches physiological levels
                                        East Kent Hospitals
                                                    NHS Trust




    Serum creatinine - measurement
•    Jaffe methods
•    Enzymatic methods
•    HPLC
•    ID-MS – reference method

    Review of the sedimentation
    process which is caused in normal
    urine by picric acid and a new
    reaction of creatinine
    By M. Jaffe (Submitted to the
    editor on 26th June 1886)
                                                     East Kent Hospitals
                                                                 NHS Trust




Serum creatinine - Jaffe method
• Cheap (1 p/test)
• Rate or end-point


                           alkaline pH
Creatinine + picric acid                 Orange-red (Janovski) complex
                                             Read @ 510 nm
In East Kent, 460,000 tests/y/620,000 popn. (2004/05)
Equates to 46 million creatinine assays/y in UK
                                  East Kent Hospitals
                                              NHS Trust




Serum creatinine - enzymatic method

      • Expensive (£0.20-£0.30/test)
      • Good agreement with ID-MS
                                                            East Kent Hospitals
                                                                        NHS Trust




     Serum creatinine - problems
Non-renal influences:
• gender, ethnicity and age
• nutrition/diet
• drugs (e.g. cimetidine)
• muscle mass

Clinical utility
• Poor sensitivity for CKD
• Not useful in ARF
• Muscle wasting disorders

Analytical problems:
• Non-specificity (protein, ketones, ascorbic acid) (pseudo-chromogens)
• Spectral interferences (icterus/lipaemia/haemolysis)
• No international standardization
            East Kent Hospitals
                        NHS Trust




   Creatinine
Standardization
    Matters
                                   East Kent Hospitals
                                               NHS Trust




     Serum creatinine
                       Analytical differences
Vitros                 persist
                       Lawson et al,
                       Ann Clin Biochem 2002

            Olympus2

            Roche
                       Serum creatinine assays
                       should be calibrated using an
         Olympus1
                       international standard
                       NKF-KDOQI 2002
East Kent Hospitals
            NHS Trust
                                                                                                           East Kent Hospitals
                                                                                                                       NHS Trust




UKNEQAS survey: accuracy against ID-MS
                                        65
                                                                     All methods
                                        60                           OCD (J&J) slides [1JJ]
 % Method Bias from Mass Spec Target.




                                        55                           Abbott reagents [11AB]
                                        50                           Bayer reagents [11TE]
                                        45                           Beckman reagents [11BK]

                                        40                           Olympus reagents [11OL]
                                                                     Roche Integra reagents [11RO]
                                        35
                                                                     Roche Modular reagents [11BO]
                                        30
                                                                     In-house reagents [14OO]
                                        25                           Olympus reagents [14OL]
                                        20                           Synermed reagents [14SR]
                                        15                           ILab reagents [6IL]

                                        10
                                         5
                                         0
                                         -5
                                        -10
                                              0   100   200     300       400        500             600   700       800
                                                              Mass Spec Creatinine (umol/L)
                                                                     East Kent Hospitals
                                                                                 NHS Trust


Calibration and population estimates
     NHANES - White Sands, Kinetic Jaffe,         MDRD – Cleveland Clinic Foundation,
     Hitachi 737 (20 umol/L>CCF)                  Kinetic Jaffe, Beckman Astra CX3




            NHANES                          -20 umol/L

                                                         Stage 3
                       Stage 3
                                                         3.2%
                       12.5%




(Clase et al JASN 2002)
                                                                              East Kent Hospitals
                                                                                          NHS Trust




                    Roche compensated assay
                     100
Difference (compensated -
 uncompensated, umol/L)



                            80                                      170/600 labs in UK use
                            60                                      these methods
                            40
                                             mean bias
                            20               + 9.8 umol/L           e.g. using MDRD for 50 y
                             0                Zero bias
                                                                    white male
                        -20                                         SCr 80 gives eGFR 94
                        -40                                         SCr 70 gives eGFR 110
                        -60
                        0         500     1000              Compensated=1.09(uncompensated) -17
                   Creatinine (uncompensated Jaffe,         Point of equivalence 185 umol/L
                                umol/L)
                             East Kent Hospitals
                                         NHS Trust




Effect of ‘compensation’ on eGFR

                             Poggio
       CKD
                             et al 2005



             kidney donors
                  Tx
                  donors
                                            East Kent Hospitals
                                                        NHS Trust




            NKDEP, September 2005
• Please Note: The National Kidney Disease Education
  Program (NKDEP) recommends that clinical laboratories
  use the current MDRD equation without changing the
  calibration of serum creatinine assays. Re-calibration of
  serum creatinine to be traceable to ID-MS reference
  methods should be introduced at the same time as a
  revised MDRD equation that has coefficients appropriate
  for use with a zero–biased creatinine method. The revised
  MDRD equation is expected in fall 2005
• (to be presented ASN November - ? Based on enzymatic
  assay).
                                           East Kent Hospitals
                                                       NHS Trust




           NKDEP, September 2005
Interim recommendations until a revised GFR estimating
  equation is available (Myers)

Implement the MDRD equation now for creatinine
  methods that are not re-calibrated to be traceable to
  IDMS

For re-calibrated creatinine methods:
Add 0.1 mg/dL (8.8 µmol/L) to the creatinine value in
  the MDRD equation to be similar to the bias that
  existed in the MDRD study lab
                                     East Kent Hospitals
                                                 NHS Trust




          NKDEP, September 2005
• Long-term recommendations
• IVD manufacturers and clinical labs should
  coordinate introduction of re-calibration to
  coincide with a revised GFR estimating
  equation based on creatinine values traceable
  to IDMS
                                East Kent Hospitals
                                            NHS Trust


        What can we do about
       standardization issues?

1) Use gold standard creatinine methods and
   an international calibrant
2) Calibrate to the MDRD „standard‟
3) Adjust the MDRD formula for our own
   assays
4) ? Do nothing pending new MDRD formula
                                          East Kent Hospitals
                                                      NHS Trust




Use reference creatinine methods
   and an international calibrant
• NKDEP laboratory working party/IFCC about to start
  working on reference materials (e.g. NIST SRM967, end
  2005)
• NKDEP LC ID-MS reference method by end 2005
• (HbA1c international standard >10 y so far)
• ID-MS methods not widely available in diagnostic labs
• Not yet suitable for routine, high-throughput use
                                                                                                  East Kent Hospitals


                      Calibrate to the MDRD ‘standard’?
                                                                                                              NHS Trust




                                 150

                                                                                   278 samples:
                                 100
                                                                                   1) Roche compensated Jaffe
Difference from ID-MS (umol/L)




                                  50
                                                                                      (Kent)
                                                                                   2) MDRD (CCF lab)
                                   0
                                                                                   3) ID-MS (Guy’s)
                                  -50


                                                                    MDRD
                                 -100
                                                                    Jaffe
                                                                                   Roche= 0.9653xMDRD -19.8

                                 -150
                                        0      500           1000           1500
                                            ID-MS creatinine (umol/L)
                                                                                        East Kent Hospitals
                                                                                                    NHS Trust




                                 Calibrate to the MDRD ‘standard’?
                                 40                                          Same dataset, limited to
                                 30
                                                                             values <150 umo/L
Difference from ID-MS (umol/L)




                                 20
                                  10
                                   0
                                 -10                                         UK laboratories unlikely
                                 -20                                         to accept ‘standardisation’
                                 -30                                         to MDRD assay
                                 -40
                                 -50       Jaffe
                                 -60       MDRD
                                 -70
                                       0    50       100       150     200
                                           ID-MS creatinine (umol/L)
                                                                                                                    East Kent Hospitals
                                                                                                                                NHS Trust




                                                                     Impact on eGFR
                                             45
                                                                                                          Using regression,
                                             40                                                           generated eGFRs for
Difference from MDRD eGFR (mL/min/1.73 m2)




                                             35
                                                                                    IDMS
                                                                                                          ID-MS and Roche Jaffe
                                             30
                                                                                    Jaffe                 assays compared to
                                                                                                          MDRD lab across
                                             25
                                                                                                          creatinine range 80-250
                                             20
                                                                                                          umol/L
                                             15


                                             10


                                             5
                                                                                                          Large differences in
                                                                                                          eGFR if calibration not
                                             0
                                                  100 90   80   70   60   50   40     30    20   10   0   taken into account
                                                       MDRD eGFR (mL/min/1.73 m2)
                                               East Kent Hospitals
                                                           NHS Trust


  Adjust the MDRD formula for our
            own assays
• Since: Roche Jaffe creatinine = 0.9653MDRD -19.8,

186 x [serum creatinine x 0.011312]-1.154 x age-0.203 x (1.212
  if black) x (0.742 if F)

• Becomes:

186 x [({serum creatinine +19.8}/0.9653) x 0.011312]-1.154
  x age-0.203 x (1.212 if black) x (0.742 if F)
                                   East Kent Hospitals
                                               NHS Trust




 Where do we go from here - 1?
• The 4v MDRD formula is being adopted slowly
  (20% of US labs)
• It improves recognition of CKD

BUT……………..
                                             East Kent Hospitals
                                                         NHS Trust




 Where do we go from here - 2?
• Its all a bit messy!
• Use of formulae has re-focused attention on creatinine
  measurement
• Creatinine standardisation is probably a long-term goal
• But an interim, revised formula is to be issued (zero-
  biased)
• Although few methods are zero-biased
• Better markers may supersede MDRD in longer-term
  (? Cystatin C ? Iohexol)
                                              East Kent Hospitals
                                                          NHS Trust




      Department of Health Working Party

•      Addressing issue of eGFR with UKNEQAS
       involvement
•      Possible areas of work:
i.     ? Develop toolkit to aid laboratories in eGFR
       implementation
ii.    ? Distribute MDRD validated samples to enable
       laboratories to generate their own MDRD adjustments
iii.   ? Monitor through UKNEQAS eGFR scheme
iv.    ? Information packs for non-specialists
v.     ? Co-ordinated implementation plan
                                             East Kent Hospitals
                                                         NHS Trust



Non-standardization issues and questions
 • Precision - ? reports should indicate CI of estimate
 • Are formulae derived in US appropriate for UK?
 • Are corrections for Black-Americans appropriate in UK?
 • What about South Asians and Chinese?
 • What about type 2 diabetes
 • Should we correct for BSA?
 • Du Bois or Haycock formula?
 • Is 1.73 m2 still appropriate? Does it matter (e.g. 1.0 m2
   for LVMI)?
 • Drug dosage – absolute or relative clearance?
 • Outcomes – BSA corrected versus uncorrected GFR
 • What about children?
      East Kent Hospitals
                  NHS Trust




END

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:34
posted:4/15/2011
language:English
pages:37