Conclusions
Apheresis with membrane differential filtration for AMD is currently undergoing a double-masked, randomized clinical trial based on the results of two small randomized clinical trials, case reports, and unpublished anecdotal evidence.  Once completed, this clinical trial may help determine if patients with non-neovascular AMD in this select group are stabilized or improved by this therapy.  Physicians should exercise caution in recommending this therapy until final results from this study are available.

Potential Benefits
If the current trial is successful, there is potential for this therapy to preserve or improve vision in a subset of patients with non-neovascular AMD who meet specific criteria.

Potential Risks
Available data suggest a 2 to 4 percent treatment-related complication rate during therapeutic apheresis.  Complications include paresthesias, hypotension, nausea, and dizziness, all of which have been minor and transient.  No serious long-term complications or side effects have been reported.  Economic risk is not assessed in this report because the charges for the treatment, if approved, are not known.

Presumed Mechanism of Action
Currently, there is no established link between either blood viscosity or the concentration of macromolecules and AMD.  However, in systemic vascular beds, there is a suggestion that macromolecules may cause endothelial cell dysfunction. ³ There are some data that suggest the hypothesis that a similar pathogenic process in the choriocapillaris may contribute to AMD. ⁴ Two papers discuss the role of vitronectin, an adhesive glycoprotein, in the formation of human ocular drusen, and they advance the theory that vitronectin plays a role in the pathogenesis of AMD. ^{5, 6}

Perfusion of tissue beds depends on capillary blood flow and endothelial cell function.  The main determinants of blood flow through a vessel are vessel diameter, pressure gradient, plasma viscosity, red blood cell (RBC) aggregation, and plasma concentration of certain blood constituents.  Altering the vessel diameter or pressure gradient may not be possible within the choriocapillaris; however, decreasing plasma viscosity and RBC aggregation might increase flow through the choriocapillaris and improve perfusion of the macula.  The possible improvement in choriocapillary blood flow and capillary endothelial cell function might enhance delivery of glucose, oxygen, and other nutrients to macular and retinal cells and facilitate removal of waste products.  This possible increased blood flow and improved perfusion may enhance functions of the retinal pigment epithelium and photoreceptor cells.  It is unknown if there are adverse effects from increasing the blood flow.

Definition of the Problem
Age-related macular degeneration is a disorder of the macula that occurs most often in patients 50 years old or older, and is the leading cause of irreversible central visual loss in Caucasians in this age group in the United States.  Using the 1990 U.S.  Census, approximately 750,000 people over 65 years of age were estimated to have severe visual impairment in one or both eyes from AMD, ⁷ the majority of whom have the non-neovascular (atrophic) form.  There is no established treatment for non-neovascular AMD.  The etiology of AMD is unknown, but there may be diverse pathogenic processes, including a genetic component. ⁸

FDA Status/Legal Status
Apheresis with membrane differential filtration for AMD is only available in the United States by enrolling in a prospective, double-masked, randomized clinical trial designed, funded, sponsored, and directed by Vascular Sciences Corporation (Tampa, Florida).  The filtration device is being tested under an Investigational Device Exemption (IDE) from the FDA.  Enrollment criteria include:

• Best spectacle-corrected visual acuity between 20/32 and 20/125 (ETDRS chart).
• Age 50 to 85 years with non-neovascular AMD with specifically defined large soft drusen.
• Elevated baseline serum concentrations of two of the following three factors: a total serum cholesterol level of greater than or equal to 200 mg/dl, a total serum fibrinogen level of greater than or equal to 300 mg/dl, or a total serum immunoglobulin A (IgA) level greater than or equal to 200 mg/dl.

Enrollment of 180 patients in one of 12 centers is underway; two centers have completed their enrollment. Follow-up for one year is mandated in the trial protocol. Preliminary results have been presented. ⁹ 

Summary of Evidence
Search Methods
A MEDLINE search from 1966 to 1999 was conducted using the key terms plasmapheresis, artificial membranes, rheology, and macular degeneration.  An EMBASE search from 1974 to 1999 was conducted using the same terms. An update search in MEDLINE and EMBASE from 1999 to 2002 was conducted in February 2003 using the same search terms.

Treatment Rationale
There is some support for the hypothesis underlying the treatment rationale in studies that have been summarized in a recent review article. ⁸ However, an observational study not included in the review article does not support this hypothesis. ¹⁰ In this study, the authors failed to find any link between atherosclerosis and its risk factors and prevalence of retinal lesions to explain the differing frequencies of age-related maculopathy between African Americans and Caucasian Americans. ¹⁰ In addition, findings from a case-control study suggest that neovascular AMD is associated with moderate to severe hypertension, but non-neovascular AMD was unrelated to hypertension or cholesterol level. ¹¹

Statistical Issues and Study Design
Results have been reported for three randomized controlled trials. ^{12, 13 ,} 9 Earlier investigations reported uncontrolled series of patients with maculopathies of different origins. ^{14, 15} Other reports on this therapy are anecdotal and/or reported in abstract form only.

Specification of Level of Evidence
Two of the randomized controlled trials ^{12, 13} are graded as Type II evidence because of uncertainty about the adequacy of the sample size and control group.  Type II evidence generally consists of randomized, controlled trials with design flaws, well-designed controlled trials without randomization, well-designed cohort or case-control analytic studies, preferably from more than one center, and multiple-time series with or without the intervention. The third randomized controlled trial report 9 was presented at the 2002 annual meeting of the American Ophthalmological Society and consists of an interim analysis of results from 43 patients; the planned total enrollment in the trial is 180 patients. As this report represents preliminary information from the trial and was not subject to the peer review process, it is rated as Level III evidence.  Level III evidence includes descriptive series, case reports, and reports of committees or organizations.

Reported Benefits
A preliminary report of the randomized controlled multicenter trial in the United States gave results of treatment in 43 patients, 28 receiving apheresis and 15 placebo-controlled patients. ⁹ These results were presented at a meeting and were not subject to the peer review process (Level III evidence). Enrollment criteria are listed above. Patients received 8 apheresis or placebo procedures over 10 weeks. At one year, the mean LogMAR line difference between the treatment group and the control group was 1.6 lines (P=0.001); 1 treatment eye had a =3-line loss of BCVA compared to 2 eyes in the control group. In a subgroup of eyes with baseline LogMAR worse than 20/40, 13 of 23 improved to 20/40 or better at one year compared to 1 of 8 in the placebo group. The changes in visual function were compared to baseline pre-treatment levels of visual acuity measured by ETDRS (LogMAR) chart and a standardized refraction and visual acuity protocol. All patients enrolled in the trial received the same daily vitamin oral supplement.
There is one study in the German literature in which 10 patients (18 eyes) received apheresis with membrane differential filtration and 10 patients (18 eyes) served as a control group. ¹²Patients enrolled had either non-neovascular or neovascular AMD.  In each group 8 out of 18 eyes showed subfoveolar subretinal neovascularizations.  The main outcome measure was best corrected ETDRS visual acuity measured by forced choice method.  The treatment group received 10 treatments over 21 weeks.  This group had a mean improvement of 1.1 (SD 1.9) lines immediately after treatment and the control group had a loss of 0.6 (SD 1.7) lines over the same period (P < 0.01).  The two groups had a difference of 0.5 lines in long term follow-up (8-30 months) without retreatment, which was not statistically significant.
A study from the same investigators reported on 40 patients, 20 receiving apheresis with membrane differential filtration, and 20 patients serving as a control group receiving no treatment. ¹³ It is not stated if this group includes patients from the earlier report. Patients enrolled had either non-neovascular or neovascular AMD.  In each group, 9 patients showed subretinal neovascularization.  The main outcome measure was best corrected ETDRS visual acuity using line assessment and the forced choice method.  The treatment group received 10 apheresis procedures over 21 weeks.  Final visual acuity measured at 24 hours following the last treatment showed a nonsignificant improvement of 0.63 lines (SD 1.8, P=0.14) on ETDRS charts while the control group had a loss of 0.94 lines (SD 1.7, P=0.02). The changes in visual function were measured in comparison with the baseline pre-treatment levels of visual acuity.

Reported Risks
Available data suggest a 4 percent immediate adverse event rate during therapeutic apheresis for nonophthalmologic indications. ^{2, 16, 17} Adverse effects reported have been minor and transient, and they include paresthesias, hypotension, nausea, and dizziness.  No serious longstanding complications or side effects have been reported.
In the case series reported by Brunner of 36 patients who received apheresis for maculopathy, five patients (14%) had symptoms of hypocalcemia because of the anticoagulant, and three (8%) had hypotensive episodes (recorded blood pressure reading of less than 90/60 mmHg). 15 In the report from the United States trial, treatment-related adverse events were observed in 2.2% (5/223) of apheresis procedures and in none of the placebo-control procedures. ⁹ No serious or longstanding treatment-related adverse events were reported.

Appropriate training for plasmapheresis procedures is necessary.

Economic risk is not assessed in this report because the charges for the treatment, if approved, are not known.

Questions for Scientific Inquiry
Several questions about apheresis for AMD need to be addressed.

• The basic hypothesis for the treatment is that improvement in microvascular blood flow and perfusion will improve visual function in patients with AMD.  There is no consensus on the reproducibility of current methods to measure ocular blood flow.  Do microvascular blood flow, perfusion, or other abnormalities play a role in the pathogenesis of AMD?
• If the current study demonstrates efficacy, at what stage of non-neovascular AMD pathogenesis will this treatment be most appropriate? Will it be effective for other forms of AMD?
• Are there long-term side effects of altering the concentration of blood constituents using this therapy?
• How long are benefits maintained? Are repeat treatments necessary? What is the ideal number and frequency of treatments to obtain or maintain any gains in visual acuity?
• How would this treatment compare with other therapies currently available or under investigation?

• Describe the scientific rationale or mechanism for action for the complementary therapy.
• Describe the methods and basis for collecting evidence.
• Describe the relevant evidence.
• Summarize the benefits and risks of the complementary therapy.
• Pose questions for future research inquiry.
• Summarize the evidence on safety and effectiveness.

1.  Malchesky PS, Skibinski CI. Summary of results of 1991 ASFA apheresis survey. American Society for Apheresis. J Clin Apheresis 1993;8:96-101.
2. Korach JM, Bussel A, Gajdos P. International forum: France. The national survey of plasma exchange and therapeutic cytapheresis in France. Transfus Sci 1995;16:363-70.
3. Lip PL, Blann AD, Hope-Ross M, et al. Age-related macular degeneration is associated with increased vascular endothelial growth factor, hemorheology and endothelial dysfunction. Ophthalmology 2001;108:705-10.
4. Harris A, Chung HS, Ciulla TA, Kagemann L. Progress in measurement of ocular blood flow and relevance to our understanding of glaucoma and age-related macular degeneration. Prog Retin Eye Res 1999;18:669-87.
5. Hageman GS, Mullins RF, Russell SR, et al. Vitronectin is a constituent of ocular drusen and the vitronectin gene is expressed in human retinal pigmented epithelial cells. Faseb J 1999;13:477-84.
6. Anderson DH, Hageman GS, Mullins RF, et al. Vitronectin gene expression in the adult human retina. Invest Ophthalmol Vis Sci 1999;40:3305-15.
7. Ophthalmology AAO. Age-Related Macular Degeneration, Preferred
   Practice Pattern. San Francisco, 2001.
8. Snow KK, Seddon JM. Do age-related macular degeneration and cardiovascular disease share common antecedents? Ophthalmic Epidemiol 1999;6:125-43.
9.  Pulido JS. Multicenter prospective, randomized, double-masked, placebo-controlled study of Rheopheresis to treat nonexudative age-related macular degeneration: interim analysis. Trans Am Ophthalmol Soc 2002;100:85-106; discussion -7.
10. Klein R, Clegg L, Cooper LS, et al. Prevalence of age-related maculopathy in the Atherosclerosis Risk in Communities Study. Arch Ophthalmol 1999;117:1203-10.
11. Hyman L, Schachat AP, He Q, Leske MC. Hypertension, cardiovascular disease, and age-related macular degeneration. Age-Related Macular Degeneration Risk Factors Study Group. Arch Ophthalmol 2000;118:351-8.
12. Widder RA, Brunner R, Walter P, et al. [Modification of vision by change in rheologic parameters in senile macular degeneration through membrane differential filtration--initial results of a randomized study]. Klin Monatsbl Augenheilkd 1999;215:43-8; discussion 8-9.
13. Brunner R, Widder RA, Walter P, et al. Influence of membrane differential filtration on the natural course of age-related macular degeneration: a randomized trial. Retina 2000;20:483-91.
14. Widder RA, Brunner R, Engels B, et al. Changes of hemorheological and biochemical parameters after plasma perfusion using a tryptophan-polyvinyl alcohol adsorber leading to clinical improvement in patients suffering from maculopathy. Blood Purif 1998;16:15-21.
15. Brunner R, Widder RA, Fischer RA, et al. Clinical efficacy of haemorheological treatment using plasma exchange, selective adsorption and membrane differential filtration in maculopathy, retinal vein occlusion and uveal effusion syndrome. Transfus Sci 1996;17:493-8.
16. McLeod BC, Sniecinski I, Ciavarella D, et al. Frequency of immediate adverse effects associated with therapeutic apheresis. Transfusion 1999;39:282-8.
17. Koga N. Efficacy and safety measures for low density lipoprotein apheresis treatment using dextran sulfate cellulose columns. Ther Apher 1999;3:155-60.
18. Ernst E RK, Mills S, et al. Complementary medicine - a definition. Br J Gen Pract 1995;45:506.
19. Fontanarosa PB, Lundberg GD. Alternative medicine meets science. Jama 1998;280:1618-9.
20. Margolin A, Avants SK, Kleber HD. Investigating alternative medicine therapies in randomized controlled trials. Jama 1998;280:1626-8.