Listening to Prozac but hearing placebo: A meta-analysis of antidepressant medication

In paper citation (Kirsch & Sapirstein, 1998)

This paper specifically looks at the effects of the placebo in placebo-controlled drug studies treating depression. This analysis only includes drug studies before 1995 and therefore misses most of the modern drugs prescribed today. However, Kirsch and Sapirstein's results are really shocking. They find that in drug studies, the placebo group improves in proportion to the effectiveness of the drug. The correlation between the placebo response and the drug response is r=.90, p<.001 and the placebo response is 75% of the drug response. So if the drug response in a sertraline drug trial was 2.0, the placebo response in that trial would be around 1.5. These numbers were slightly higher for active placebos... side effects actually may convince people that they are receiving full treatment, and therefore cause them to expect a better response. This is also interesting in light of the more modern Turner and colleagues meta-analysis, which found that 49% of FDA approved clinical trials show no significant difference between placebo and antidepressant trials. Perhaps both the experimental and control patients are improving in proportion to the expected efficacy of the drug, and so only the most effective drugs will come out ahead of their placebos.

Furthermore, the authors compared the effect of taking a placebo pill to receiving no treatment while on a wait-list for psychotherapy trials. For the people on wait-lists, or receiving no placebo treatment at all, their depression scores worsened overall during the course of the study. This shows that the placebo pill is more effective than no treatment.

My remaining questions actually involve the rating scales of depression. It could be that different research centers treat their patients more holistically, and that is why the scores are so well correlated, but it could also be that people respond to the depression interviews in a rote manner. For example, if I know the symptoms of depression, and I want to be a good research subject, I may describe my pre-treatment symptoms in a way so that the doctor will give me the quick fix and treat me with an antidepressant. Then when the doctor asks me follow ups, I also know what improvement should look like, and I've been looking out for that in my daily life, and so I focus on the improvement for my follow up questionnaire. Whether the drug has actually improved my mood is hard to tell, because I started by looking for symptoms to describe to my doctor, and then I started looking for improvements to tell to my doctor. Thus the placebo effect.

Does anyone else think that is totally plausible? I guess Kirsch & Sapirstein would, but they do not mention it in their discussion. In cases of mild to moderate depression, I would like to see a study that looks at various types of placebos to see what works best. Inert substances should be better for our bodies in the long run, and especially useful for pregnant mothers who desire treatment for mood disorders.

Selective Publication of Antidepressant Trials and its Influence on Apparent Efficacy

In paper citation (Turner, et al., 2008).

In groundbreaking work, Turner and his colleagues invoked the freedom of information act to "check up" on the efficacy of antidepressants. Basically, the majority of studies with negative or questionable results about the efficacy of an antidepressant versus a placebo never get published and therefore do not have a chance to influence people's mentality about the efficacy of antidepressant drugs.

Basically, among the published literature, 94% of antidepressant trials conducted are positive. However, among the FDA registered clinical trials of antidepressants, only 51% of antidepressant trials are positive. This gives us great pause for consideration as pharmaceuticals make billions of dollars a year on these drugs.

I personally would trust the FDA effect sizes now that they are out. The effect size compares the difference in means between the control and the placebo group at the end of the treatment period divided by the pooled standard deviation.

The overall effect size was .31 for the modern antidepressants.

Animal models of Depression

In paper citation (Deussing, 2006).

I have been curious about what sorts of animal models of depression exist because I am already thinking about developing experiments related to depression and sleep.

First of all, I am thoroughly dissatisfied by the two most common methods for assessing despair in rodents: the forced swim test and the tail suspension test. This is because they are acutely sensitive to the administration of antidepressants, and we know that antidepressants typically take about four weeks to display therapeutic effects. Furthermore, when suspending a male rodent from his tail, he receives a huge surge of testosterone from his testes because this is the area manipulated by the tail-suspension test. Tail suspension is very abnormal, and so are testosterone surges.

Also, do humans show an immediate decrease in despair after taking antidepressants? I feel like I would have heard about it if this were the case, but I have not. I have heard that more people commit suicide in those first few weeks, but I have never heard this directly linked to acutely increased motivation (like the motivation to swim or struggle in the rodent tests). We should have a better human correlate than the motivation to kill oneself.

My three favorite tests of depression are the sucrose preference test, the novelty-induced hypophagia test, and the dexamethasone suppression test. The sucrose preference test measures a rodent's motivation to seek rewarding stimuli (such as sugar in the water) and is sensitive to chronic antidepressant treatment. The novelty-induced hypophagia measures a rodent's anxiety level by presenting the rodent with a conflict. The rodent is faced with a desirable food item inside of a novel environment. It can either avoid a novel environment, or enter the novel environment and consume the food. The biggest problem with this paradigm is that it measures anxiety, not depression, and some rodents never choose to consume the novel food. The dexamethasone suppression  test looks at the ability of a synthetic glucocorticoid to downregulate glucocorticoid expression.This basically measures the neuroendocrine health of the rodent.

In terms of creating depression in a rodent, my three favorite methods are via social stress, maternal deprivation in the first two weeks of life, and olfactory bulbectomy. Social stress is most like human stress, but is not as reproducible as learned helplessness paradigms. Similarly, early life stress is less reproducible, and has not been tested with the typical antidepressants yet (which I find shocking!). Finally, the olfactory bulbectomy is easiest to do in mothers shipped from the rat factory.  However, I am not so certain how to make rat mothers depressed and not just stressed. Is it even possible to isolate the two? Time will tell.

Sleep Disturbance Associated with an enhanced Orexinergic system induced by chronic treatment with paroxetine and milnacipran

In paper citation (Rahmadi, et al., 2011).

Mice treated with the SSRI paroxetine 10mg/kg or SNRI milancipran (30mg/kg) had their sleep recorded on day 7 and their orexinergic receptor and mRNA for orexinergic receptors quantified on day 28. Both antidepressants significantly decreased total sleep time and decreased total NREM sleep. OX1R and OX2R mRNA expression significantly increased in the hypothalamus of animals on antidepressants, as did the H1R and histidine decarboxylase mRNA expression in the frontal cortex.

Paroxetine still has an affinity for the H1R receptor of .0045*10^-7/equilibrium dissociation constant, which may be blockading the H1R receptor and thus upregulating it (Richelson, 1996). Richelson did not have the binding affinity for milnacipran in his paper.

Histidine decarboxylase is involved in the synthesis of new histamines, so the blockade of H1R receptors might promote this increased synthesis, and thus increased mRNA expression in the frontal cortex.

Orexins are synthesized in the lateral hypothalamus, so the increase in Orexinergic receptor mRNA expression might indicate that little orexin is making it back to the hypothalamus for use in a feedback mechanism. The orexins might be binding in wake-promoting areas and getting metabolized there. Increasing total orexins should increase a person's wakefulness.

REM: Neurobiological mechanisms for the regulation of mammalian sleep-wake behavior: Reinterpretation of historical evidence and inclusion of contemporary cellular and molecular evidence

In paper citation (Datta & MacLean, 2007)

Historical perspective

• Jouvet's transection studies in the 1960s showed that any cut rostral to the pons eliminated REM sleep signs in the forebrain.
• The model in the 1970s, termed the reciprocal interaction model, assumed that the aminergic inputs were REM-off and that cholinergic inputs were REM-on (and thus off until REM sleep initiation).
• Major locus for the cholinergic cells shifted from the mPRF to the PPT as more research showed REM-on Ach cells in the PPT.

Cellular-Molecular Network model of sleep regulation

REM sleep sign generators:

• MRF and the medullary magnocellular nucleus regulate the cortical EEG readings in REM.
• LC-alpha neurons create muscle atonia
• Peri-abducens reticular formation creates rapid eye movements (also responsible for horizontal saccades in awake individuals)
• Caudo-lateral peribrachial area creates PGO waves (waves that begin in the pons, propogate to the lateral geniculate body and the occipital cortex). The p-wave occurs only in REM, it is glutamatergic and excitatory, and it tends to bursts during REM as well as having a high tonic rate. 
• Pontis oralis controls the hippocampual theta rhythm
• parabrachial nucleus controls that brain and body temperatures and autonomic fluctuations. Autonomic system suspend its regulatory functions throughout REM sleep. More people die of heart attacks during REM than any other time of day.

REM is turned on when aminergic cells are markedly reduced and cholinergic cell activity is relatively high. (0:0.65 ratio)

Regulation of cholinergic tone in the REM sleep-sign generators:

• PPT contains 3 types of neurons: REM-on, wake-REM-on, and state independent.
• REM-on and wake-REM-on cells increase activity at the initiation of slow wave sleep.
• Cholinergic cells do not fire in a bursting manner.
• Activation of only kainate receptors in the LDT and PPT results in 65% activated cells, whereas activation of NMDA receptors will activate 100% of the cholinergic cells in the PPT resulting in wakefulness.
• Kainate has a lower activation threshold than NMDA, but both are glutamate receptors.
• The influx of calcium ions activate adenylyl cyclase, creating cAMP, which phosphorylates PKA.
• GABA can prevent this process because GABA receptors couple to Gi/Go g-proteins which inhibit adenylyl cyclase and prevent the cAMP-PKA signal transduction pathway. 
• Kainate receptors desensitize quickly, but increased cytosolic PKA can phosphorylate GluR6 and modulate channel function for 3-25 minutes. 
• Thus PKA may be responsible for sustaining activity in the PPT cells.
• Axons of the PPT and LDT terminate in the mPRF in the cat.
• Glutamated in the PPT induces REM sleep.
• Ach release in the mPRF increases in REM sleep.
• Open question if mPRF is the  effector zone, or if there are multiple effector zones.

Regulation of monoaminergic tone in the REM sleep-sign generators:

• Serotonin
• The raphe nucleus contains a significant proportion of cells containing GABA and other neurotransmitters.
• 5HT-ergic cells project to the sleep sign generators. 
• 5HT-ergic cells stop firing during REM sleep. 
• Decreasing 5-HT in the brain increases the appearance of PGO waves.
• Increasing 5-HT in the brain can block PGO waves without stopping REM sleep. 
• Norepinephrine
• Neurons of the LC decrease their firing rate at sleep onset and remain completely silent until 5-10 seconds before the beginning of wakefulness.
• Cooling the LC induced tS-R within 3 minutes and REM within 4 minutes.
• Mechanisms for regulating NE and 5HT REM-off cell activity
• 3 hypothesis:
• GABA inhibits the RN and LC
• a pacemaker mechanism
• withdrawl of histaminergic and hypocretinergic tone
• GABAergic mechanism
• Some of the LC and RN neurons are GABAergic; these neurons could be inhibiting the rest  of the structure endogenously
• Level of spontaneously produced GABA in the LC and RN is maximum during REM sleep and minimum during wakefulness. 
• Single cell recording shows a reciprocal neural discharge between 5HT and GABA cells in the DRN.
• Pacemaker mechanism
• buildup of GABA turns off  cAMP-PKA signal, shutting off LC
• DRN then controlled by LC activity.
• NE antagonists will turn off the DRN
• Withdrawl of histaminergic and hypocretinergic excitatory tone
• LC and DRN receive projections from the HA cells in the PH and Hcrt cells in the LH.
• LH and PH are less active during REM than during wakefulness. 
• HA can prevent the cessation of DRN REM-off cells during REM sleep.
• REM-off cells are also in the ventrolateral periaqueductal gray (vlPAG) and the lateral pontine tegmentum (LPT)
• vlPAG seems to be more involved in pain than sleep, though.

Remaining questions:

• To where do PPT cells project? Only the mPRF?

SWS: Neurobiological mechanisms for the regulation of mammalian sleep-wake behavior: Reinterpretation of historical evidence and inclusion of contemporary cellular and molecular evidence

In paper citation (Datta & MacLean, 2007)

There are two major warrants to the activity dependent homeostatic theory of sleep initiation:

1. "The waking state requires a critical level of brain activity, which is maintained by a steady flow of ascending impulses arising in the brainstem reticular formation"
2. "A reduction of tonic activity of the ascending reticular system (ARAS) is responsible for physiological sleep."

Basically, the initiation of sleep is a passive process caused by the withdrawl of wakefulness. But what causes this withdrawl. Here's what the theory says:

• During wakefulness, the increased rate of metabolite synthesis is higher than the rate of clearance. Individual cells begin to demand a lower metabolic state and cease firing. Eventually, this effects behavior at the systemic level, and the body initiates sleep. 
• This theory predicts that duration of sleep periods are inversely correlated with the rate of metabolite clearance.
• Factors only involved in sleep induction are not metabolic byproducts. These are synthesized after SWS has started.
• Adenosine
• Cells use ATP by catabolizing it into adenosine and ADP. Adenosine moves along its concentration gradient and can therefore get built up by periods of large use. 
• Adenosine accumulates in the BF and cortex during forced sleep deprivation.
• Direct administration of adenosine increases sleep duration and enhances SWS activity in the rat.
• Blocking adenosine synthesis eliminates SWS and increases wakefulness.
• Adenosine also inhibits the activity of Hcrt neurons in the LH.
• Inhibitory amino acids (GABA)
• GABA - everywhere
• glycine - spinal cord & brainstem
• Glutamate is decarboxylated to form GABA by GAD. 
• Increased neuronal activity results in a local increase in GABA synthesis.
• Glycine synthesized from the degradation of serine. 
• Choline can metabolize into glycine by stepping through betaine and losing  its methyl groups.
• A global increase in GABA via ventricular infusion promoted the physiological signs of NREM sleep. 
• Prostaglandin
• Prostaglandins are a naturally occurring unsaturated fatty acid group (loosely) made from arachidonic acid.
• PG is observed most often in the CSF between the arachnoid membrane and the pia mater.
• Primarily synthesized in the leptomeninges, the epithelial cells of the choroid plexus, and the oligodendrocytes.
• Sleep deprivation raises the typical PG concentration in the CSF.
• Injection of PGD2 into the Preoptic area, or into the lateral vesicles increases NREM sleep.
• Cytokines
• Cytokines stimulate subtle changes in cellular metabolism.
• Cytokines affect the input-output relationships within their neural circuit of origination.
• Interleukine-1beta is highest when the demand for NREM sleep is highest in the rat (beginning of the day), taper off for awakening
• In humans, IL-1B is highest at initiation of sleep, and lowest at awakening.
• Injecting IL-1B into the brain directly increases amount of NREM sleep, but only if the the subject is already in NREM sleep. (NOT causative)
• Substances that inhibit IL-1B decrease spontaneous sleep. 
• TNF-alpha is also 10X higher at the initiation of sleep than during its minimal waking values.
• TNF-alpha in the POA enhances NREM sleep in rats.
• TNF inhibitors also inhibit NREM sleep and decrease spontaneous sleep. 

Mechanisms for the generation and maintenance of SWS:

• After sleep initiation, GABA and galanin synthesizing cells in the anterior hypothalamus/ POA become active and project to the major wake promoting areas, inhibiting them. 
• GABA hyperpolarizes the thalamus to raise the threshold for sensory information getting relayed to the cortex.
• POA critical for SWS generation. Lesions in the POA can prevent SWS in mammals.
• FMRI data shows that mPOA is more active than other parts of the hypothalamus and basal forebrain during SWS. 
• POA lesions will knockout SWS, but only for a matter of days depending on the extent of the damage. 
• Paradoxically, NE and 5HT in the POA increase POA activity and can induce wakefulness.
• GHRH (a sleep inducing factor) may be needed to help GABA from the POA work. 
• GHRH made in the arcuate nucleus (largest #), ventromedial nucleus (VMN), and paraventricular nucleus (PVN). These cells project to the anterior pituitary (from the arcN) and the POA.
• GHRH  is high around the initiation of sleep, and highest immediately after sleep.
• Highest GHRH synthesis occurs at the period of deepest SWS. 
• 2/3 of all growth hormone (GH) secreted in young males occurs during SWS. 
• GHRH injections systemically and ventricularly increase SWS in rats.
• GH injection decreases SWS by providing negative feedback to the GHRH production system.
• GHRH  is released into the POA and binds to GHRH recptors to activate POA GABAergic cells.

Remaining questions: 

• Does TNF-alpha injection only increase NREM sleep during NREM periods, or can it initiate NREM?
• What are the sleep induction factors that get synthesized during SWS?

WAKE: Neurobiological mechanisms for the regulation of mammalian sleep-wake behavior: Reinterpretation of historical evidence and inclusion of contemporary cellular and molecular evidence

In paper citation: (Datta & MacLean, 2007)

This is my first review article, so I will do my best to lay out the information in a concise and organized manner. I will not include citations, because the paper has 17 pages of citations and should therefore be used to confer all of these facts.

Historical background on consciousness & identification of sleep stages

• Pre-12th century belief was  that sleep is a passive unconscious state.
• Early hindu civilization distinguished two types of sleep: prajna, meaning dreamless sleep, and taijasa, meaning dreaming sleep.
• EEG was first recorded in 1924, and published in 1929.
• In 1953 Aserinsky discovers REM sleep in infants and discovered that it also existed in adults.
• REMs lasted as long as 50 minutes and started 90-120 minutes after sleep onset.
• Dement observed the first REM and NREM sleep in the cat in 1958. (Although Klaue first looked at sleep in the cat in 1937) This was the animal model of choice until the late 1900s. 

Physiological characteristics or wake, NREM and REM

• Polysomnography is the combination of EEG measurements (brain), EMG measurements(muscles), and EOG measurements (eyes) to study sleep.
• While we are awake, we display an activated EEG (20-60 Hz), muscle tone, and voluntary movements/ progressive and logical thoughts.
• There are 4 stages of NREM sleep that correlate to 2 states of SWS in animals. 
• In stage 1 sleep, our eyes might still be open, but the cortex has low voltage, 3-7 Hz oscillations called vertex sharp waves of activation. This really isn't seen in animals. 
• Stage 2 sleep is most like SWS-1 in animals. This stage is characterized by sleep spindles in the cortical EEG. In humans, up to 80% of stage 2 sleep might actually be tSR (transitional REM) sleep in animals.
• Stage 3 & 4 are the deepest stages of NREM sleep, equivalent to SWS-2 in animals. This is considered deep sleep, or delta sleep, due to the presence of high-amplitude, low-frequency (.1-4Hz) waves in the cortival EEG.
• REM sleep is characterized by:
• Low amplitude, high frequency waves in the cortical EEG
• Atonia (no muscle tone/activation)
• singlets and clusters of rapid eye movements
• theta rhythm activity in the hippocampal EEG (hard to measure in humans)
• spiky field potentials in the pons (p-waves)
• Humans have 4-6 sleep cycles a night. The period lengths of each REM-NREM sleep epoch increases with brain size across species (Cats have longer cycles than rats).

Regulation of sleep timing:

• Organisms remain active during hours when the opportunity to acquire food exceeds the risk of predation, but they sleep during times when the need for vigilance is minimized. 
• The suprachiasmatic nucleus  has many circadian clock genes that encode different types of proteins that act as transcription factors to regulate their own transcription.

Wake-promoting systems of the brain:

• Five cell types in the ascending reticular activating system are reponsible for promoting wakefulness. These are:
• Noradrenergic (NE) cells in the locus coeruleus (LC)
• Serotonergic (5-HT) cells in the raphe nuclei (RN)
• Cholinergic (Ach) cells in the pedunculopontine tegmentum (PPT)
• Glutamatergic (Glut) cells in the midbrain
• Dopaminergic (DA) cells in the substantia nigra compacta (SNc)
• NE cells of the LC:
• fire maximally during wake behavior and steadily decrease until they cease firing during REM sleep.
• become active immediately prior to spontaneous wakefulness, suggesting an anticipatory role in wake-behavior.
• Experimental application of NE in the thalamo-cortical, hypothalamo-cortical, and basalo-cortical activating systems induces cortical activation and promotes wakefulness.
• Mice lacking NE fall asleep more rapidly after a mild stress.
• 5-HT cells of the RN:
• fire maximally during wake behavior and steadily decrease until they cease firing during REM sleep
• do not anticipate wake behavior and are therefore probably an effect, not a cause of wakefulness.
• BUT: Lesions of the RN  increase wakefulness and decrease SWS.
• Direct application of 5-HT to the preoptic area resulted in a decrease in wakefulness allowing SWS.
• Ach cells of the PPT
• PPT neurons also synthesize nitric oxide, a gaseous neuromodulator, that regulates wakefulness by controlling activity levels of PPT cells. 
• Electrical stimulation of PPT promotes wakefulness.
• Four major types of cholinergic cells in the PPT: REM-on, W-REM-on, Wake-on, and sleep unrelated.
• PPT at 100% activation during wake, 65% activation during REM, and 7.4% activation during SWS.
• Ach cells may be in a tug of war with the 5-HT cells to stay active. When Ach drops out therefore, 5-HT could induce SWS.
• Midbrain reticular formation
• Electrical stimulation of the MRF is a reliable technique for inducing cortical activation in rats & cats.
• Kainic acid and glutamate microinjections in the MRF causes arousal in the cat and rat.
• Activity in MRF of humans is higher during wake than SWS. 
• DA cells of the SNc and VTA
• Do not display robust alterations in firing rate across sleep-wake states
• DA cells will burst during REM sleep.
• Extracellular DA is significantly elevated during wakefulness. 

• Wake promoting cell groups outside the ARAS:
• histaminergic (HA) cells in the posterior hypothalamus (PH)
• hypocretin-containing (Hcrt) cells in the lateral hypothalamus (LH)
• cholinergic (Ach) cells in the basal forebrain (BF)
• cells in the SCN
• Histaminergic cells in the PH
• Patients with encephalities lethargica showed a prolonged period of sleepiness with a higher waking threshold. This was the result of an injury between the PH and the rostral midbrain. 
• The tuberomammilary nuclei (TMN) contains HA-ergic cells that project diffusely throughout the brain, and specifically to wake-promoting structures. 
• Single cell recording shows that these neurons are active during wakefulness and silent during sleep, but their activity precedes and predicts awakening.
• Inhibiting TMN via GABA suppresses wakefulness.
• Drugs that enhance HA signalling increase wakefulness.
• Mice without histidine decarboxylase can't even stay awake in a novel environment.
• Hypocretinergic cells in the LH
• Hcrt-containing neurons also co-express glutamate and pentrazxin. 
• Hcrt neurons are more active during wakefulness than SWS.
• Hcrt increases arousal via an excitatory effect on wake-promoting systems in the brain.
• Selective lesioning of Hcrt neurons in the LH increases SWS and REM and decreases wakefulness
• Hcrt knockout mice look like narcoleptics.
• Wake-promoting function of LH neurons could be associated with motivated behaviors.
• Cholinergic cells in the BF (contains 4 cholinergic nuclei)
• BF Ach cells receive inputs from other brainstem and hypothalamic wake-promoting systems and project to the cerebral cortex.
• Single cell recording studies show that BF cells are more active during wakefulness than SWS.
• Cells in the SCN
• SCN cells fire more frequently during wakefulness than sleep. 
• Lesion to the SCN has had mixed effects, sometimes decreasing wakefulness, and sometimes having no effect. 
• Mutated Bmal1 and Cry1/Cry2 genes in the SCN increase NREM sleep at the expense of wakefulness. 
• PFC/mPFC
• 4 main subdivisions of mPFC:
• dorsal contains the medial agranular & anterior cingulate cortex; implicated in motor behaviors
• ventral contains the prelimibic cortex & infralimbic cortex; implicated in emotional cognitive and mnemonic processes
• 3 main subdivisions of PFC (primates):
• orbital -emotional behavior
• medial - emotional behavior
• lateral - executive functions
• failure to fall asleep due to "racing thoughts" is a cognitive flexibility failure, caused by hyperactivity of the mPFC/PFC.
• PKA pathways are disinhibited with age; increased PKA activity disrupts cognitive flexibility.